INFORMATION TO USERS This reproduction was made from a copy o f a docum ent sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this docum ent, the quality o f the reproduction is heavily dependent upon the quality o f the material subm itted. The following explanation o f techniques is provided to help clarify markings or notations which may appear on this reproduction. 1. The sign or “target” for pages apparently lacking from the docum ent photographed is “ Missing Page(s)” . If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure complete continuity. 2. When an image on the film is obliterated with a round black mark, it is an indication o f either blurred copy because o f movement during exposure, duplicate copy, or copyrighted materials th at should n o t have been filmed. For blurred pages, a good image o f the page can be found in the adjacent frame. If copyrighted materials were deleted, a target note will appear listing the pages in the adjacent frame. 3. When a map, drawing or chart, etc., is part o f the material being photographed, a definite m ethod o f “ sectioning” the material has been followed. It is customary to begin filming at the upper left hand com er o f a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again-beginning below the first row and continuing on until complete. 4. For illustrations that cannot be satisfactorily reproduced by xerographic means, photographic prints can be purchased at additional cost and inserted into your xerographic copy. These prints are available upon request from the Dissertations Customer Services Departm ent. 5. Some pages in any docum ent may have indistinct print. In all cases the best available copy has been filmed. University M iarinlm s International 300 N. Zeeb Road Ann Arbor, Ml 48106 8407203 K a c z m a r, Sw iatoslav W olodym yr PART PER TRILLION DETERMINATION OF 2,3,7,8-TETRACHLORODIBENZO PARA DIOXIN IN MICHIGAN FISH PH.D. M ic h ig a n State University University Microfilms International 1983 300 N. Zeeb Road, Ann Arbor, Ml 48106 Copyright 1980 by Kaczmar, Swiatoslav Wolodymyr All Rights Reserved PLEASE NOTE: In all c a s e s this material has been filmed in the best possible way from the available copy. Problems encountered with this d o cum ent have been identified here with a check mark V 1. Glossy photographs or p a g e s ______ 2. Colored illustrations, paper or p rin t______ 3. Photographs with dark b ack g ro u n d ______ 4. Illustrations a re poor c o p y ______ 5. P ag es with black marks, not original 6. Print shows through a s th ere is text on both sid e s of p ag e______ 7. Indistinct, broken or small print on several p a g e s 8. Print exceeds margin req u irem en ts______ 9. Tightly bound copy with print lost in spine______ 10. . copy___ Computer printout p ages with indistinct print_____ 11. P ag e(s)____________ lacking w hen material received, and not available from school or author. 12. P ag e(s)____________ seem to b e missing in numbering only a s text follows. 13. Two pages n u m b e re d _____________. Text follows. 14. Curling and wrinkled p a g e s ______ 15. O ther______________________________________________________ __________________ University Microfilms International PART PER TRILLION DETERMINATION OF 2,3,7,8-TETRACHLORODIBENZO-P-DIOXIN IN .MICHIGAN FISH By Swiatoslav Wolodymyr Kaczmar A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Fisheries and Wildlife and Center for Environmental Toxicology ABSTRACT PART PER TRILLION DETERMINATION OP 2,3,7,8-TETRACHLORODIBENZO-P-DIOXIN IN MICHIGAN FISH By Swiatoslav Wolodymyr Kaczmar Samples of Carp (Cyprinus carpio) and Sucker (Catostomus commersoni) were collected from major rivers in Michigan and analyzed for the presence of 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD) by a procedure optimized for the isomer-specific quantification of 2,3,7,8-TCDD at the ng/kg level. Reference standards of 21 TCDD isomers were synthesized by controlled pyrolysis of selected chlorophenols. Residues of 2,3,7,8-TCDD ranging from 17 to 586 ng/kg were detected in fish from various portions of the State, indicating that 2,3,7,8-TCDD is more widely distributed in Michigan than previously thought. To my parents TABLE OF CONTENTS Page I. I N T R O D U C T I O N ........................ 1 II. TOXICOLOGY OF THE D I O X I N S .......................... 4 A. Acute Effects 4 B. Sub-acute Effects C. Metabolism of D i o x i n s ............................ 14 D. Teratogenicity of T C D D ............................26 .................................. ................... . . . . . 6 E. Mutagenicity and Carcinogenicity ............... 30 F. Human Effects of T C D D ............................ 36 III. ENVIRONMENTAL INPUTS AND FATE OF T C D D ............... 38 IV. A. Chlorophenol Manufacturing ...................... 38 B. TCDD Release During Industrial Accidents . . . . 4^ C. Product Contamination .......................... 4j? 1. Phenoxy acid h e r b i c i d e s ...................... 4“ 2. Agent O r a n g e .................................43 3. Pentachlorophenol .......................... 45 D. Formation of Dioxins During Combustion ......... E. Environmental Dynamics and F a t e ..................59 F. The Dioxin Controversy in M i c h i g a n ............... 64 4® E X P E R I M E N T A L ............................................70 70 A. Purpose of Research B. Introduction to Analytical Problem and Methodology...................................... C. Detail of Procedure for TCDD A n a l y s i s ........... 79 D. Synthesis and Determination of Resolution of TCDD I s o m e r s ....................................... 94 1. E x p e r i m e n t a l ................................... 98 2. GC/MS evaluation of pyrolysis products . . 3. R e s u l t s ...................................... 100 E. Quality Control/Quality A s s u r a n c e ...............*3.2 1. Method v a l i d a t i o n ........................... **2 2. Reagent p u r i t y ................................ 115 3. Maximization of selectivity towards 2,3,7,8-TCDD 118 F. Results of 2,3,7,8-TCDD Determinations ......... 118 TABLE OF CONTENTS (cont'd) Page V. D i s c u s s i o n .........................................126 A. Evaluation of Analytical Technique .......... 126 B. Geographical Distribution ofResidues. C. Ecological and Public Health Significance of the Residues............................... 139 . . . 132 VI. CONCLUSIONS.........................................146 VII LITERATURE CITED ................................. VII. A P P E N D I X ........................................... 164 148 LIST OF TABLES Page 1. Single-dose L D cq Values of 2,3,7,8-TCDD in Various Mammalian Species ........................ 5 2. Sub-Acute Effects of 2,3,7,8-TCDD............. .. . 8 3. Metabolism and Pharmacodynamics of 2,3,7,8-TCDD . 15 4. Toxicity of Various Stereoisomers of 2,3,7,8-TCDD in the Guinea P i g ............................... 25 5. Teratogenicity Studies with 2,3,7,8-TCDD ......... 6. TCDD-Induced Mutagenicity Studies................ 32 7. 2,3,7,8-TCDD Carcinogenesis Studies.............. 34 8. Toxic Effects of TCDD Reported in M a n ............ 37 9. Dioxin Levels Associated with the Incineration of Municipal Waste................................... 48 10. Results of a 2,3,7,8-tetrachlorodibenzo-p-dioxin Fish Monitoring Program Conducted by the U.S. Environmental Protection Agency in Michigan During 1978...................................... 27 66 11. Excerpts of Data from the Trace Chemistries of Fire Report........................................68 12. Products and Yields of Chlorophenate Pyrolysis . . 103 13. Capillary GC Retention Times of the 22 TCDD's . . . 104 14. Optimum Working Recoveries Determined Using 14C - 2 ,3,7,8-TCDD .................................. 114 15. Procedural Blanks................................. 117 16. Fish Sample L o g ................................... 120 17. Results of Fish Analyzed for 2,3,7,8-TCDD 18. 2,3,7,8-TCDD in Livers of Mink Fed Carp from Saginaw Bay ....................................... 141 19. TCDD Residues in Fish Before and After Broiling . . . . 122 . 143 LIST OF FIGURES Page 1. Dibenzodioxin .......................................... 1 2. Formation of 2,3,7,8-TCDD During Commercial Synthesis of Chlorophenols .......................... 3 3. The Cytosolic Binding Affinity and Biological Potency of Dibenzo-p-dioxin Congeners Relative to 2 , 3 , 7 , 8 - T C D D ........................................ 23 4. Routes of Formation of Chlorinated Dioxins During Combustion ..................................... 5. Flow Diagram of the Analytical P r o c e d u r e ............. 75 6 . Reaction Pathways of 2 ,3,6-trichlorophenate ......... 7. 54 96 GC/MS Chromatograms of the TCDD's Produced by Pyrolysis of C h l o r o p h e n a t e s .......................... 105 8 . Map of Sample S i t e s ................................... 125 1 I. INTRODUCTION Chlorinated dibenzo-p-dioxins are a group of chemical compounds which are among the most hazardous pollutants in the environment. These compounds, collectively referred to as dioxins or PCDDs, are tricyclic aromatic compounds which can occur as 75 unique congeners differing by extent and sites of chlorination. The basic dibenzo-p-dioxin molecule is nearly planar and has eight possible sites of chemical substitution (Figure 1). Fig. 1. Dibenzo Dioxin. It can be chlorinated at any of the ring positions with the exception of the oxygens at points 5 and 10. There are two possible monochlorinated isomers (MCDD), 10 dichlorinated (DCDD), 14 trichlorinated (T^CDD), 22 tetrachlorinated (TCDD), 14 pentachlorinated (P^CDD), 10 hexachlorinated (HgCDD), two heptachlorinated (H^CDD) and a single octachlorinated isomer (OCDD). Chlorinated dioxins are extremely stable, non-reactive compounds. The 2,3,7,8-TCDD isomer is resistant to the action 2 of concentrated acids and bases and maintains its chemical stao bility at temperatures as high as 700 C. In contrast to the chemical inertness of the dioxins, they are relatively photolabile and have been demonstrated to photolyze in the environ­ ment with typical half-lives of 2-3 hours. TCDD does not undergo hydrolysis in water and is resistant to microbial attack. The 2,3,7,8-TCDD isomer exhibits an extremely low vapor pressure (1.5 x 10 —5 mm Hg), is soluble in water to only 0.2 ug/1 and has an estimated octanol/water partition coefficient of 1 x 10 . The environmental dynamics of chlorinated dioxins are similar to some of the longer lived chlorinated hydrocarbons. PCDDs can be synthesized by direct halogenation of the un­ substituted dioxin using chlorine gas and a suitable catalyst. They can also be prepared by pyrolysis of ortho substituted chlorinated phenol salts. Both methods produce dioxins in high yield, but the latter reaction allows a greater degree of con­ trol over the specific end products. Although a few of these compounds have been synthesized in the laboratory, they are not intentionally produced for any commercial purpose, but arise as impurities in materials synthesized by treating chlorinated benzene at elevated temperature and pressure under alkaline conditions (Figure 2). This synthetic step is utilized in the production of several commercially important products, notably; the fungicide pentachlorophenol (PCP), the germicide hexachlorophene and the herbicide 2,4,5-trichlorophenoxyacetic acid (2,4,5-T). PCDDs have been detected in significant amounts in all of these products. NaOH a ow Fig. 2. Formation of 2,3,7,8 TCDD during commercial synthesis of chlorophenols. PCDDs have also been detected in fly ash and extracts from a number of combusted materials. It is theorized that their presence is the result of reactions occurring at trace levels during combustion. Because chlorophenols are very im­ portant chemical intermediates and combustion is relied upon as a means of waste disposal and energy conversion, PCDDs could be widely distributed in the environment. However, due to the extreme analytical sophistication required for trace dioxin analysis, very few residue data are available with which to evaluate the environmental implications of dioxin-yielding processes. In this study an accepted technique for dioxin analysis is verified and its chromatographic resolution is extended to specifically detect the 2,3,7,8-TCDD isomer. Samples of fish from 19 major rivers in Michigan are analyzed for 2,3,7,8-TCDD, providing data on the residue levels and geographical distribu­ tion of 2,3,7,8-TCDD in the state. 3 4 II. TOXICOLOGY OF THE DIOXINS A. Acute Effects PCDDs, particularly the 2,3,7,8-TCDD isomer, are of special concern due to their extremely high toxicity. The 2,3,7,8-TCDD isomer is perhaps the most toxic small molecule known to man. Its single-dose L D 5Q to various mammals is on the order of a few micrograms per kilogram of body weight. There is a marked variation in species sensitivity to 2,3,7,8-TCDD, whose LI>5 q ranges from 0.6 ug/kg in the guinea pig to over 3,000 ug/kg in hamsters (Table 1). Administration of 2,3,7,8-TCDD does not cause immediate death; but acts slowly, affecting progressive weight loss and debilitation. Death usually occurs more than 3 weeks following administration of a fatal dose. Histologic examination at necropsy reveals multiple organ damage. Although the exact mode of action of TCDD has not been delineated, its activity appears to be linked to its affinity towards a specific cytosolic receptor (1). The receptor is a protein that binds with TCDD, then travels to the nucleus and to a structural gene locus") (which has in mice been termed the "Ah initiating a pleiotropic response resulting in the in­ duction of a number of coordinately expressed and possibly re­ pressed critical proteins or enzymes. Interestingly, this receptor has been demonstrated to exert varying degrees of affinity towards a number of halogenated aro­ matic hydrocarbons which include the chlorinated dibenzo-p-dioxins, dibenzofurans, azo(xy)benzene, naphthalenes and biphenyls and 5 Table 1. Single-Dose L D ™ Values of 2,3,7,8 TCDD in Various Mammalian Species. Animal and Treatment Observations Guinea pig, oral single dose LD™(male) = 0.6 ug/kg LD^J: (female) = 2.2 ug/kg 50 Guinea pig (female) 1 ug/kg once a week, orally: LD 100 after 24-32 days Reference (2 ) (3) Guinea pig, oral single dose: LD 9 q = 3 ug/kg (4) Guinea pig (female) single oral dose: LD5q = 2 ug/kg (5) Monkey, oral: L D 50 = 70 ug/kg (6 ) Rat, intraperitoneal adult male: L D ™ = 60 ug/kg weanling male: L D ™ = 25 ug/kg 50 3-methylcholanthrene pre-treated weanling LD = 4 4 ug/kg 50 male: adult female: (7) L D ™ = 25 ug/kg 50 L D ™ = 50 ug/kg *50 (8 ) Rat, oral: male female L D ™ = 22 ug/kg LD^Jj = 45 ug/kg (2 ) Rabbit, oral: dermal: L D ™ = 115 ug/kg LDgjj = 275 ug/kg (2 ) Rat, oral: Mouse, intraperitoneal C57BL/6J: L D ™ = 132 ug/kg DBA/2J: L D ™ = 620 ug/kg B6D2F1/J: LDgJJ = 300 ug/kg (9) Mouse, C57/B1, oral: L D 50 = 1 2 6 ug/kg (10 ) Mouse, oral: L D 50 = 114 ug/kg (11 ) Hamster, oral: LD[.q = 5,051 ug/kg (12) Hamster, intraperitoneal: oral: L D ™ = 3,000 ug/kg LDgjj = 1,157 ug/kg (13) 6 the brominated biphenyls. Whereas the patterns of toxic response and species sensitivity are common amongst certain halogenated aromatic hydrocarbons, the dose required to achieve a particular degree of toxicity is related to the material's measured affinity for the receptor. The 2,3,7,8-TCDD congener has been shown to have the highest affinity for the receptor and the lowest ED j q * The degree of affinity of the other active congeners appears to be a function of their ability to act as a planar stereoisomer Of B. 2,3,7,8-TCDD. Sub-acute Effects Administration of 2,3,7,8-TCDD elicits a wide range of toxic responses, none of which is thought to be independently responsible for the ultimate death of the test animal. TCDD poisoning can be described as a debilitating process induced by pathological changes in a number of organs and systems. There are marked species, strain and sex related differences in observed L D ^ ' s and patterns of toxic response to TCDD. Table 2 lists the observations of a number of investigators conducting research into the subacute effects of 2,3,7,8-TCDD. Neglecting species differences, the most common responses include: 1) a wasting syndrome: a progressive weight loss which may not be related simply to decreased food consumption; 2) skin disorders: acneiform eruptions or chloracne, alopecia, edema, hyperkeratosis and blepharitis due to hyper­ trophy of the Meibomian glands; 7 3) lymphoid evolution: atrophy of the thymus and spleen causing humoral and cell-mediated immunosuppression and/or bone marrow and hematological dyscrasias; 4) hepatotoxicity: a) hepatomegaly due to hyperplasia and hypertrophy of parenchymal cells with a proliferation of the smooth endoplasmic reticulum, b) general or localized necrosis, c> lipid accumulation. 5) disordered porphyrin metabolism: porphyria cutanea tarda; 6) endocrine and reproductive dysfunction: altered plasma levels of steroid and thyroid hormones with a) menstrual irregu­ larities, reduced conception rate, early abortion, excessive menstrual and post-conceptional hemorrhage and anovolution in females and b) testicular atrophy and decreased spermatogenesis in males; 7) teratogenesis: 8) carcinogenesis. kidney malformations and cleft palate In addition to 2,3,7,8-TCDD, most of these effects have been elicited by other dioxin congeners as well as by certain polychlorinated dibenzofurans, polychlorinated biphenyls and polybrominated biphenyls. The commonality of responses further suggests that these materials are acting by the same mechanism. Table 2. Sub-acute effects of 2 , 3,7, 8 TCDD Animal and Treatment Observations Reference Rat, oral, 50 ug/kg Hypertrophy of liver and regression of thymus. Rat, oral, 0 .1-10 ug/kg A single dose of 0.1 ug/kg (14) increased liver weights. Daily doses at 10 ug/kg caused death of all animals within 2-4 days, there was an accumulation of large amounts of lipid in the liver, accompanied b^.an increase in incorporation of C leucine into liver protein. Rat, oral, single dose, 5 and 25 ug/kg Three days after dosing, changes in (15) endoplasmic reticulum, especially of the bile duct. Regeneration after 20 days. Rats, oral, single dose 200 ug/kg Depressed food intake and weight loss. Mean time to death was 20 days. Gastric hemorrhagy and jaundice. Pathologic changes in polyneucleated liver parenchymal cells. (8) (16) Influence on the lymphoid organs, (17) Rats, 0.1-25 ug/kg especially thymus. Depletion of Mice, 1.0-50 ug/kg lymphoid cells in spleen and lymph Guinea pigs, 0.0008-3.0 nodes. Liver changes less in mice ug/kg. All orally and guinea pigs than rats. At sublethal doses, degenerative livers, megalocytosis and unusual numbers of multinucleated giant hepatocytes. Changes in cells of renal collecting tubule and in thyroid follicles. Rats, oral, single dose. Three days after dosing, lesions of (is) plasma membrane associated enzymes. Complete loss of ATPase reaction in parenchymal cells of the centrilobular zone. Changes in 6 -nucleo­ tidase and acid phosphatases were secondary to the parenchymal cell damage. 9 Table 2. cont'd. Rats, oral; 0.001, 0 .01, 0.1 and 1.0 ug/kg 5 days a week for 13 weeks 1.0 ug/kg: mortality, icterus, (19) increased serum bilirubin and alkaline phosphatase. Porphyria, liver damage, thymic atrophy and functional suppression of repro­ ductive organs. 0.1 ug/kg: Decreased appetite and body weight. Liver degeneration and lymphoid depletion. Porphyria and increases urinary a -ALA.in females. Depression of packed cell volume, RBC and hemoglobin in males* 0.01 and"0.01 ug/kg: increased liver weights. 0.1 ug/kg/day: increased incidence (20) Rats, dietary, of carcinomas and increased mor­ 0 .001, 0 .01, 0.1 ug/kg tality, decreased weight gain, por­ daily for 2 years. phyria, increased A-ALA and serum alkaline phosphatase, a-glutamyl transferase and SGPT. Morphologic changes in hepatic, lymphoid res­ piratory and vascular tissues. 0.01 ug/kg/day: same as higher dose, but to a lesser degree. 0.001 ug/kg/day: No observable effects in males. Increased incidence of swollen hepatocytes in females. Rats, dietary, 0 .001- 0.1 ug/kg/day for 90 days prior to mating; Three-generation reproductive study. (21) 0.1 ug/kg/day: significant decreases in fertility and neonatal survival in F. generation. 0.01 ug/kg/day: significant decreases in fertility in F. and F 2 generations but not in F fl. Decreases in litter size, gestation survival and neonatal survival and growth. 0.001 ug/kg/day: no effects on repro­ ductive capacity, survival or litter size and weight i n •any generation. Rats, dietary, 50-1,000 ppb in feed Atrophy of the thymus and spleen, (22) dilation of common bile ducts, gastrointestinal hemorrhage, micro­ scopically observed severe liver necrosis, cellular proliferation in the bile ducts, cellular proliferation in the bile ducts, cellular hyper­ plasia of the mucosa and edema in the common bile ducts; decreased spermatogenesis. 10 Table 2. cont'd. Rats, Pre- and postnatal maternal exposure at 5 mg/kg Suppression of thymus-dependant functions for 6 months. (23) Mice, oral 125 ug/kg single dose Weight loss and death with fatty (10) liver after 21 days. Progressive necrotic centrilobular liver lesions. Mice, oral 0.5-20 ug/kg/week 4 weeks. Dose schedules of 1 ug/kg or more, (24) followed by Salmonella infection, resulted in significant increases in mortality and decrease in the time from infection to death. The most notable finding is that low levels of TCDD, which do not produce clinical pathological changes, have the capacity to affect host defense. Mice, oral, female, 10 ug/kg/day 10-14 days after application; depres- (25) sion of the blood platelets by antiboldy reaction. Diminished clot retraction is consistent with the observed thrombocytopenia. The num­ ber of erythrocytes, their cell vol­ ume and the number of neutrophils was increased. Mice, oral, 0.2-25 ug/kg/wk 2 or 6 weeks. The application of 1, 5, or 25 ug/kg (11) caused an increase of liver and thymus weights. Total neutrophils were increased. Hemoglobin, serum protein and globulins were signifi­ cantly decreased after 6 doses of 25 ug/kg/wk. TCDD was porphyrogenic. Mice, oral 1 and 10 ug/kg/day Increase of SGOT and serum cholesterol,(26) decrease of blood glucose, increase of SGOT, SGPT, lactate dehydrogenase and bilirubin in serum. Mice, oral 10 and 100 ppb in feed for 5 weeks. 100 ppb: marked suppression of total (27) serum protein, gamma globulin and albumin; increase in beta globulins. 10 ppb: reduction in the primary and secondary antibody response to teta­ nus toxin, sheep erythrocytes and lowered resistance to challenge with Salmonella tvphimurium or Listeria monocytogenes. 11 Table 2. cont'd. Mice and rats, 500 ug/kg Tubule damage in kidneys, fragmentation of cell membranes, fatty degeneration of the liver. (28)> Monkeys, rats and chickens, oral. In all species, hypoplasia of the (29) lymphatic tissue and bone marrow. Ascites, hydrothorax and hydropericard. In addition, rats developed gastric ulcers and enlargement of liver. Monkeys additionally showed alopecia, hyperkeratosis and a decrease of the number of spermatocytes and spermatides. Male rats, 25 ug/kg Rabbits, 25 ug/kg Guinea pigs, 5 ug/kg intraperitoneally. Liver function was assessed 10 days (30) following administration. Rats: decreased ouabain clearance unaltered indocyanine green clearance elevated serum SDH, unaltered .SGPT, GGPT? reduced bile flow, salt excre­ tion and erythritol clearance. An 8% re­ duction in body weight. Rabbits: unaltered ouabain clearance; greatly reduced ICG clearance; ele­ vated SDH and SGPT; unchanged GGTP • 2% reduction in body weight. Guinea pigs: unaltered ICG clearance; unchanged SDH, SGPT, and GGPT; 35% reduction in body weight. Guinea pigs, oral, 0.008 to 1.0 ug/kg/wk for 8 weeks. Mouse, oral, 0.2-25 ug/kg Rats, dermal, 0 .2-5.0 ug/kg Guinea pigs: up to 0.2 ug/kg caused (3) loss in weight, lymphopenia and atro­ phy of the lymphatic organs. All higher doses were lethal. Mice: Weight loss and thymic atrophy. Rats: Weight loss and reduction of weights of thymus and adrenal gland. No skin sensitivity. Guinea Pig, i.p. 2 ug/kg single dose Marked hyperlipidemia; 19-fold increase (31) in VLDL; 4-fold increase in LDL. Female Rhesus Monkey, 50 ppt in diet for 20 months. Four abortions and 1 stilbirth following(33> breeding attempts at 7 months with a cumulative dose of 0.36 ug/kg. Inability to conceive. Hair loss, hyperkeratosis, weight loss. Decrease in serum cholesterol, hematocrits and white blood cells. Increased SGPT. 12 Table 2. cont'd. Female Rhesus Monkeys 500 ppt in diet for 9 months. (32) Anemia within 6 months. Pancytope­ nia after 9 months. Death in 5 of 8 between months 7 and 12 at a to­ tal exposure of 2-3 ug/kg. Exten­ sive hemorrhage, hypocellularity of the bone marrow and lymph nodes. Hypertrophy, hyperplasia and meta­ plasia of the epithelium in the bronchial tree, bile ducts, pan­ creatic ducts, salivary gland ducts, palpebral conjunctivae. Squamous cell metaplasia and keratinization of the sebaceous glands and hair follicles in the skin. Female Rhesus Monkeys Single oral dose 70 ug/kg Significant weight loss, blepharitis, (6) loss of fingernails and eyelashes; ^ facial alopecia with acneiform erup­ tions: mild anemia, neutrophilia, lymphopenia? decreased serum chole­ sterol; increased serum triglycerides. Hyperplastic and metaplastic changes in sebaceous glands of eyelid, ear canal and epithelial hyperplasia in the renal pelvis, stomach, gall blad­ der and bile duct. Liver normal. Rabbit, oral single dose, 1 ug/kg Serious liver damage and chloracne. (34) Irreversible toxic effects and death (35) Fish* Silver salmon and guppy within 10-80 days of exposure. Inhibited growth and caused death of ca. 23 mg/g for 24 hr. the trout after 33 days exposure. Rainbow trout, 6.3 ug/wk/10 animals Fishi Induced retardation of embryonic develPike and rainbow trout, opment-and growth. Dose-related inFreshly fertilized eggs, cidences of hemorrhages, edema and yolk sac fry and hepatic injury, followed by death. juveniles Survivors showed skeletal malforma­ 0.1 to 100 ppt in tions, inclusion bodies in stomach, tap water pancreas and liver. Fish; guppies, 0 .1 , 1.0 and 10.0 ppb for 120 hrs (36) Caused 100% mortality within 37 days post­ exposure. Toxicity delayed at least 5 days. Mean survival time = 21 days. (38 13 Table 2. cont'd. Fish: Coho Salmon 96 hrs. at 0.54 ng/g and 5.4 ng/g. Guppies, 96 hrs. at 0.08 ng/g and 0.8 ng/g. Coho Salmon: 0.54 ng/g; no observable (37) effects up to 60 days postexposure. 5.4 ng/g: growth and survival re­ duced over 114 days postexposure. Guppies: 0.08 ng/g; no observable effects. 0.8 ng/g; developed fin disease 42 days postexposure. Fish: model ecosystem Mosquito fish died at all exposure 0.1, 1.0, 10.0 mg/kg levels. Death accompanied by nasal applied to sediment. hemorrhagy. Snails, algae and water water concentration was fleas were unaffected. 1 .0 , 1 0 .0 , and 100.0 ng/1 at 3 days. Hens, oral 9-60 ug in "toxic fatty material" (39) "Chick edema disease " (hydropericardium, hydroperitoneum) with high mortality. ABBREVIATIONS: A-ALA = A-aminolevulinic acid RBC = red blood cells SGPT = serum glutamic pyruvic transaminase SGOT = serum glutamic oxaloacetic transaminase SDH = sorbitol degydrogenase ICG = indocyanin green VLDL = very low density lippoprotein LDL = low density lippoprotein (40) 14 C. Metabolism of dioxins • An important aspect of the toxicology of 2,3,7,8-TCDD is its metabolic disposition in the host. Table 3. lists the re­ sults of a number of investigations conducted into the pharmaco­ dynamics of TCDD. Two types of metabolism studies are considered: those dealing with the tissue levels and excretion rates of TCDD, and those concerned with the extent and nature of its metabolites. Studies into the tissue distribution and excretion of 2,3,7,8-TCDD indicate that most of the TCDD administered to a test animal partitions into the liver, with lesser amounts asso­ ciated with adipose tissue. The lowest levels are found in skeletal muscle, heart, testes, blood and the brain. A notable exception is the rhesus monkey, which absorbs most of the ad­ ministered TCDD into skin and skeletal muscle, with less than 10% of the dose retained in the liver (42). Measured whole-body half-lives appear to be species depen­ dant, and range from 11 to 30 days. Excretion occurs primarily via feces, with smaller amounts removed with urine. of Metabolism 2,3,7,8-TCDD was initially thought to play a minor role in the excretion of TCDD, since a large amount of the administered dose was observed to be excreted in the feces in an unmetabolized form. Tulp and Hutzinger (47) isolated mono- and dihydroxy meta­ bolites of dibenzo-p-dioxin, 1-MCDD, 2-MCDD, 2,3-DCDD, 2,7-DCDD, 1,2,4-TrCDD and 1,2,3,4-TCDD from feces and urine of rats. Sulfur-containing metabolites, possibly glutathione or mercapturic acid conjugates, were also detected. They demonstrated 15 Table 3. Metabolism and Pharmacodynamics of TCDD. Animals and Treatment Observations Reference Rats; 14C-TCDD oral, 50 ug/kg At 21 days: 56% of activity ex(8) creted in feces; 4.5% in urine; remainder present mainly in the liver, associated with microsomes. Rats: oral in feed 7 and 20 ppb for 42 days Excretion T,= 12 days in males and (41) 15 days in females. At day 42, 85% and 70% of the total do.se was still retained in the livers of males and females, respectively. H-TCDD Rats and monkeys 400 ug/kg Liver retention was 40% in rats (42) and 5% in monkeys, who partitioned the TCDD into skin, muscle and fat while rats partitioned the TCDD mainly into liver. Rats: oral A large portion of the activity was (29) associated with the microsomal fraction of the liver. 14 C-TCDD Rats: oral 50 ug/kg Two days following treatment: 30% (43) excreted in feces; 45% localized in liver; 10% localized in fat. Over 21 days: 53, 13 and 3% of the dose was excreted via feces, urine and expired air. Rats: oral dietary as OCDD Octachlorodibenzo-p-dioxin (OCDD) (44) present unmetabolized in feces, liver and adipose tissue. None detected in other tissues or urine. 14 C-TCDD Pregnant rats: intraperitoneal Minor (<1%) quantites found in pups following maternal exposure. Rats: oral various congeners 250 mg/kg Demonstrated that primary metabolism (47) takes place as hydroxylation at the 2-, 3-, 7-, or 8- position exclu­ sively via an arene oxide intermediate. Demonstrated the presence of sulfurcontaining conjugates. 14 C-TCDD Male rats: oral 15 ug/kg 2, 4, or 6 doses. (45) Bile contained at least 5 distinct (48) polar metabolites. 8-glucuronidase treat­ ment verified metabolites as glucuronide conjugates. 16 Table 3. cont'd. 14 C-TCDD Rats: oral T,= 5.6 to 17.4 days. (46) 53.2% eliminated via feces 13.2% via urine 3.2% via respired air liver and fat contained 10 times more activity than other tissues. 14 C-TCDD Rats: oral single or repeated doses of 1 ug/kg 5 days/wk. 22 days following single application, liver and fat contained levels 50 and 10 times higher than other tissues. Excretion T,= 30 days. 99% of the activity excreted via feces. A body burden of 74% of the steady-state reached within 7 weeks with repeated dosing. Mouse, rat and rabbit: 14 C-TCDD Excretion via feces. TCDD persists unmetabolized in liver, concen­ trated in the microsomal fraction. No metabolites observed in treated microsome preparations from mouse, rat or rabbit liver. C-TCDD Guinea pig: gavage single dose Excretion T,= 22 to 43 days. (51) 22 days following dosing; fat, liver, adrenal and thymus, con­ tained 0.75, 0.40, 0.33 and 0.72 %dose/g respectively. Plasma, urine and bile contained no activity. C-TCDD Guinea pig: i.p. 20 ug/kg Partitioned into liver microsomes (52) and fat. Also found in skin. T^= 30 days. 94% in feces, 6% in urine. 14C-TCDD Pish: 0.54 ng/kg 96 hrs Excretion T,= 107 days. Beef Cattle 24 ppt TCDD in feed 28 days At day 28: 85 ng/kg in fat, 8 ng/kg in liver, 7 ng/kg in kidney, 2 ng/kg in muscle. Excretion T, from fat = 16.5 weeks. 9 5% steady -State reached in 499 days with a maximum steady-sate level of 594 ng/kg when fed 24 ppt in feed daily. (49) (50) (37) (68) 17 Table 3. con t ’d. 14 C-TCDD Hamster; i.p. 650 ug/kg TCDD in liver and adipose tissue (53) exclusively in unmetabolized form. T. = 11 days 59% excreted in feces 41% in urine Activity in urine and feces due to metabolites, possibly glucuronide conjugates. No parent TCDD found in urine or bile. C-TCDD Mouse; i.p. 10 ug/kg C57B1/6J: Tj= 17 days 74% in feces 26% in urine DBA/2J 37 days (54) 70% in feces 30% in urine B6D2F1/J: ^ = 1 7 days 73% in feces 27% in urine Both parent (25-45%) and metabolites extracted from feces. 14 C-TCDD Rats; i.p. 500 ug/kg 14 Urinary and biliary activity strictly due to metabolized TCDD. Metabolites unique from those found in guinea pigs or mice. (54) C-TCDD Guinea pigs: i.p. 500 ug/kg Urinary and biliary excretion activity (54) strictly due to metabolites of TCDD. Observed metabolites different from those extracted from mouse or rat urine. JH-TCDD Isolated hamster hepatocytes. Observed at least 4 polar metabolites (55) corresponding to those previously isolated from hamster urine and bile. Some of the metabolites were glucuronide conjugates. Isolated rat hepato­ cytes from rats pretreat'ed with: TCDD (5 ug/kg i.p.) phenobarbital (PB) CoCl2 (60 mg/kg s.c.) Hepatocyte preparations from TCDD (55) or PB- treated animals showed in­ creased metabolite formation while C o C ^ r SKS-525A or metapyrone treat­ ment reduced metabolism, indicating that TCDD is metabolized by cytochrome p450 monooxygenases. Some hepatocytes were also treated with: SKF-525A (O.lmM) metapyrone (0.5 mM) 18 Table 3. con t ’d. Rats: Naturally ocurring age and sex related (7) and TCDD-induced differences in MFO activity were inversely correlated with the 20-day o f TCDD. Suggests that TCDD induces its own metabolism. H-TCDD Rats: female Sprague-Dawley Various formulations Monitored uptake by liver as a func(56) tion of formulation. Oral: TCDD in 50% ethanol = 36.7% Aqueous suspension of soil and TCDD = 16% Aqueous suspension of activated carbon + TCDD = 0.07% Dermal: Methanol + TCDD = 14.8% Vaseline + TCDD = 1.4% Polyethylene glycol + TCDD = 9.3% Soil/water paste + TCDD = 1.7% Activated carbon/water + TCDD paste = 0.05% 3H-TCDD Beagle: 1 year old Four 1.4 mg doses Oral intubation Collected bile. (58) no glucuronide conjugates bile extract contained 5 identifiable metabolites: 2,3,6,8-tetrachloro-7-hydroxydibenzop-dioxin: 2,3,8-trichloro-7-hydroxydibenzo-p-dioxin ? tetrachlorodimethoxydiphenylether; l,2,-dichloro-4,5,-dimethoxybenzene. Guinea pigs: oral TCDD metabolites iso lated from dog bile administered at 0 .6 , 6.0 and 60 ug/kg to the guinea pigs The bile extract TCDD-metabolites were 100 times less toxic than 2,3,7,8 TCDD. (59) 3H-TCDD Isolated rat hepatocytes: Incubated isolated hepatocytes' with TCDD. Identified metabolites as glucuronide conjugates. Primary metabolites extracted following treatment with -glucuronidase identified as l-hydroxy-2 ,3,7,8tetrachlorodibenzo-p-dioxin and 8-hydroxy-2 ,3,7-trichlorodibenzop-dioxin. (57) MFO = mixed function oxidase 19 that primary metabolism of dioxins takes place exclusively at the 2-, 3-, 7-, or 8- positions via 2,3 epoxides. Since these positions are blocked by chlorine in 2,3,7,8-TCDD or OCDD, they concluded that metabolism of these two congeners by this route should not readily take place. However, significant amounts of polar metabolites of 2,3,7,8-TCDD have since been detected in urine and bile of treated animals (48, 53, 54, 58, 59) and in extracts from treated cultured rat hepatocytes (55, 57). with Incubation of the metabolites -glucuronidase indicated that some of the metabolites may be glucuronide conjugates (48, 53, 55, 57). Metabolites isolated from rat hepatocyte cultures treated with ar-giucuronidase were identified as 1-hydroxy-2,3,7,8-TCDD and 8-hydroxy2,3,7-TrCDD (57), while extracts from bile of a 2,3,7,8-TCDD treated dog were found to contain 5 major identifiable non­ conjugated primary metabolites: 2,3,6,8-tetrachloro-7-hydroxy- dibenzo-p-dioxin; 2,3,8-trichloro-7-hydroxydibenzo-p-dioxin; an unassigned trichloro-dihydroxydibenzo-p-dioxin; a tetrachlorodihydroxydiphenylether and 4,5-dichlorocatechol (58). The major metabolite from the dog, 2,3,6 ,8-tetrachloro-7hydroxydibenzo-p-dioxin suggests metabolism through an arene oxide intermediate via an NIH shift. Induction of mixed function oxidase (MFO) enzyme systems increased metabolite formation, suggesting that 2,3,7,8-TCDD is metabolized by cytochrome p-450 containing monooxygenases (55). Animals with lower basal MFO activity exhibited lower LD^ q values while L D j q S were increased in TCDD or 3-methylcholanthrene (3-MC) 20 induced individuals (7). There is also some correlation observed between species sensitivity towards TCDD poisoning and rates of elimination and metabolism of 2,3,7,8-TCDD. For example the guinea pig, which exhibits an LD^q of 2 ug/kg and a measured excretion half-life of 30 days, excretes 94% of the applied dose in the feces in the unmetabolized form (52); while the hamster, which exhibits an of greater than 3,000 ug/kg and an ex­ cretion half-life of 11 days, excretes 40% of the applied dose as a metabolite in urine (53) . The dog is another example of an animal which has a high 2,3,7,8-TCDD i*D5Q and considerable ability to metabolize 2,3,7,8-TCDD (58). Although the relation­ ship is not clearly consistent among all of the species studied, the capacity to metabolize and rapidly excrete 2,3,7,8-TCDD may in part contribute to and partially explain differences in spe­ cies susceptibility towards TCDD toxicity. Studies conducted into the metabolism and biochemical re­ sponses elicited by TCDD have also shown that 2,3,7,8-TCDD is an extremely potent stimulator of enzymes associated with the metabolism of xenobiotics. TCDD has been demonstrated to induce aryl hydrocarbon hydroxylase (AHH), 7-ethoxycoumarin-O-deethylase, hepatic p-nitroanisole-o-demethylase and 3-methyl-4-methylaminoazobenzene-N-demethylase. TCDD has the ability to induce cyto­ chrome p-450 in liver, bowel, lung, kidney and skin (60) . The 2,3,7,8-TCDD isomer is the most potent inducer of AHH and cyto­ chrome p-450 known. It is 30,000 times more potent than 3-MC as as inducer of hepatic hydroxylase activity in the rat (60). The implications of the induction of these enzyme systems 21 are varied. On the one hand, an induced individual .is capable of metabolizing and excreting xenobiotics at a faster rate, thereby reducing the time and extent of exposure and concomit­ tant manifestations of toxicity. However, a number of toxic compounds and carcinogens require metabolic activation. An individual in an induced state can activate these materials faster and to a greater extent, thereby increasing their toxic potential. Another implication of the induction of metabolizing enzymes is their increased ability to degrade or produce endo­ genous materials, causing severe metabolic imbalance. example, For aminolevulinic acid synthetase activity is increased following TCDD administration, causing severe disruption of heme biosynthesis resulting in porphyria cutanea tarda. The toxicity of 2,3,7,8 TCDD may be directly linked to the induction of enzymes which it instigates. However, this relationship is not clear, since studies into the effects of enzyme induction by 3-MC have failed to reproduce the toxic manifestations elicited by TCDD. Due to the similarities between enzyme systems induced by TCDD and those induced by polynuclear aromatic hydrocarbons such as benzo-a-pyrene and 3-methylcholanthrene (3-MC), induction by 2,3,7,8-TCDD has been classified as 3-MC type induction. Studies by Poland (61) have shown that 3-MC type induction is a receptor-mediated process which is initiated by binding to a specific cytosolic receptor. This single receptor is believed to be common for most materials eliciting 3-MC type induction. Binding to this receptor is a competitive process and the relative 22 potency of various PCDD congeners for lethal, teratogenic and acneigenic effects has been demonstrated (62) . Those congeners which do not evoke a response in test animals have a very mild or nonexistent affinity for the receptor while.those of. inter­ mediate to high binding affinity, such as 1,2,3,4,7,8-HgCDD and 2,3,7,8-TCDD are of medium and high toxicity, respectively. Figure 3 presents data of Poland (61) which demonstrates the correlation between binding affinity of various PCDD con­ geners to mouse hepatic cytosol and their AHH-inducing ability in the chick embryo. The observed trends are supported by observations made by McConnel (5), who reported that the 2-, 3-, 7-, and 8- positions must be chlorinated to achieve the greatest degree of toxicity. Addition of chlorine atoms at the ortho position reduces toxicity, but not as severely as the deletion of a chlorine at one of the lateral positions. A study conducted by Bradlaw (63) investigated the com­ parative induction of AHH activity in rat hepatoma cell cultures by various analogues of dibenzo-p-dioxin. The trends they ob­ served are consistent with the requirements outlined by McConnel. They found the following congeners, listed in order of decreas­ ing potency towards AHH induction, to have greater than 0.2% of the activity of 2,3,7,8-TCDD: 2,3,7,8-TCDD, 2,3 ,7,8-TBrCDD, 1,2,3,4,7,8-HgCDD , 1,2,3,7,8-PgCDD / 1,2,3,4,7 P^CDD , 1,3,7,8TCDD / 1,2,3,6,7,8-HgCDD7 1,2,3,7,8,9-HgCDD , 1,2,3,4,6,7,8-E?C D D . Congeners which induced less than 0.2% activity were: 2,3,7TrCDD , 1,2,3,8-TCDD , 1,2,3,4,6,7,9-H?CDD , 1,2,3,6,7,9-HgCDD. Inactive congeners tested were dibenzo-p-dioxin, 1,3,-; 1,6-; RELATIVE BINDING AFFINITY 3808R RELATIVE BIOLOGICAL POTENCY 100t 100 167 100 43 43 20 22 16 8 13 3 14 006 tt RELATIVE BINDING AFFINITY 600 .■to t : : tractive |5 4 • KTV inactive | 9 4 * I0 '|" •Cl -Ct inactive (2 7 ? tO '| inactive (9 4 *10*1 ■Cl inactive ( 5 4 -tO’1) inactive (9 4* t o ' ) inactive |5 4 * t 0 * j inactive (4 7*1 0 ', | tractive (54*10 *| inactive ( 9 4 * 1 0 ') inactive |2 7 *IO » | inactive (9 4 * I0 '| inactive (11* to -' 1 inactive ( 9 4 * I0 '| $06 Cl RELATIVE BIOLOGICAL POTENCY Cl Cl Ch j N m . ::608c Cl Cl Figure 3. The Cytosolic Binding Affinity and Biological Potency of Dibenzo-p-Dioxin Congeners Relative to 2,3,7,8-TCDD. (From Poland et al. Ref. 61) 24 2,3-; 2,7-; and 2,8-DCDD, 1,2,3,4-TCDD and 1,3,6,8-TCDD and 1,2,4,6,7,9-HgCDD. Table 4 lists a number of halogenated aromatic hydrocarbons, including dioxins, which evoke identical patterns of toxic re­ sponse in the guinea pig but differ in potency. stereoisomers of 2,3,7,8-TCDD. They are all The observed order of reactivity in this table are consistent with the theory that the molecule must be nearly planar and have molecular structures that roughly o fit into a rectangle 3 x 10 A with halogen atoms in the four corners of the rectangle (65). Additional halogens at the ortho positions fall outside the rectangle and may, in some cases, not seriously effect receptor binding. However, if they de­ stabilize the overall polarizability of the carbon-halogen bonds in the molecule, they can effect its attraction towards the polar receptor matrix and decrease its binding affinity and ulti­ mate biological potency. A very interesting observation was made by Cheney (66) on the basis of regression analysis of the affinity of a number of 2,3,7,8-TCDD congeners and their potency for AHH induction. It was determined that lateral substituents do not always affect binding and induction of AHH to the same degree. this can be seen in Figure 3. Examples of Cheney reasoned that the dis­ crepancy could be the result of a decrease in the effective dose caused by metabolism or excretion prior to reaching the receptor, but he theorized that the receptor could also re­ quire activation subsequent to binding. It was observed that the lateral substituents effect binding, but also must function to "turn on" the receptor to an active state which proceeds to 25 Table 4. Toxicity of Various Stereoisomers of 2 , 3 , 1 ,8-TCDD in the Guinea Pigk Name of Compound LD^ q (ug/kg) a 2 . 3 . 1 . 8-Tetrachlorodibenzo-p-dioxin 2 1 . 2 . 3 . 1 . 8-Penatachlorodibenzo-p-dioxin 3 2 . 3 . 1 . 8-Tetrachlorodibenzofuran 7 2.3.4.7.8-Pentachlorodibenzofuran <10 2 . 3 . 1 . 8-Tetrabromodibenzofuran <15 1 . 2 . 3.4.7.8-Hexachlorodibenzo-p-dioxin 73 1.2.3.7.8.9-Kexachlorodibenzo-p-dioxin 60-100 1.2.3.6 .7.8-Hexachlorodibenzo-p-dioxin 70-100 2.3.4.6 .7.8-Hexachlorodibenzofuran 2.3.7.8-Tetrachlorofluorenone 120 >100 1.2.4.6 .7-Pentabromonaphthalene 200 2.3.6.7-Tetrabromonaphthalene 242 1.2.3.4.6 .7-Hexabromonaphthalene 361 3,3',4,4'-Tetrachlorobiphenyl <552 1.2.3.4.6 .7.8-Heptachlorodibenzo-p-dioxin >600 1.2.4.7.8-Pentachlorodibenzo-p-dioxin 1125 2,3,3',4,4',5,5'-Heptachlorobiphenyl >3000 3,3' ,5,5'-Tetrafluoro-4,4'-dichlorobiphenyl >3000 2.3.6.7-Tetrachloronaphthalene >3000 1.2.3.5.6 .7-Hexabromonaphthalene >3610 3,3',4,4',5,5'-Hexachlorobiphenyl 2.3.7-Trichlorodibenzo-p-dioxin 223 29 ,444 aHartley strain guinea pig given a single oral dose (gavage) and observed for 30 days From Mckinney and McConnell,(Ref.64) 26 initiate a series of events resulting in AHH induction and numerous other responses. There are stereoisomers which can be shown to bind, but do not appear to "turn on" the receptor very efficiently. A lateral substituent that was observed to have this capacity is the N0 2 group which, when in a configura­ tion out of the lateral plane of the dioxin molecule, functions to facilitate binding to the receptor but does not activate it. According to this theory, a compound could be devised which would bind to the receptor with a very high affinity, but not turn it on, thereby acting as an antidote for chlorinated hydro­ carbon poisoning of the 2,3,7,8-TCDD type. Blocking the acti­ vity of the receptor would also prove very valuable in deter­ mining its function in the normal activities of the cell. D. Teratogenicity of TCDD In addtition to highly fetotoxic and reproductive effects, 2,3,7,8-TCDD has been shown to be a very potent teratogen. Table 5 lists the results of a number of studies performed which examined the teratogenic potential of 2,3,7,8-TCDD. Ad­ ministration of 2,3,7,8-TCDD to mice, rats, hamsters, rabbits and monkeys during gestation induced various malformations, including: Fetal cleft palate, hydronephrosis, marked hemor- hage, edema, extra ribs and the absence of eyelids. The tera­ togenic effect most commonly observed was fetal cleft palate in mice. It appears that 2,3,7,8-TCDD is the most potent teratogen observed to date. teratogenic effects in mice. Doses as low as 1 ug/kg cause Poland has demonstrated that the 27 Table 5. Teratogenicity Studies With 2,3,7,8 TCDD. Animal and Treatment Observation Chick embryo: injection of 0.9 ug into yolk sack Toxic effects to the embryos, resuiting in 40% hatchability and pronounced embryopathies. (40) Hamster, oral 2,4,5-T with 0.1-45 ug/kg TCDD Absence of eyelids; hemorrhages and marked edema. (67) Rats: oral 500 mg/kg p.c. days 13 to 15 of gestation Fatty inclusions in embryonic livers. (68) Mouse: oral 1-3 ug/kg Strains: CD-I, DBA/2J and C57BL/6J All three strains developed cysti- (69) nephrotic kidneys and cleft palate in offspring. Rats: oral 0 .5 ug/kg Cystinephrotic kidneys in pups. Mice: oral 3 ug/kg C57BL/6J Application during pregnancy caused (70) fetal cleft palate and cystinephrotic kidneys. Postnatal maternal exposure caused hydronephrosis in foster pups. Rats: oral single dose of TCDD (5 ug) ‘* c - Mice: oral 1 ug/kg/day Mice: oral 20-50 ug/kg in the interval 7-13 days of pregnancy ~ ~ Reference (69) A very high activity was present in (71) the fetuses, especially in liver. Application from day 6 to 15 of preg- (72) nancy caused an increase in fetal mortality and cleft palate. A single dose in this interval caused (72) the same teratogenic effect without fetocidal influence. The teratogenic maximum was by dosing at day 1 1 . Rats: oral Teratogenic no-effect level 0.03 ug/kg (73) After application of 0.125-2.0 ug/kg, Sprague-Dawley 0.03-8.0 ug/kg/day there was a increase in the number days 6-15 of gestation of resorptions and dead fetuses, inte­ stinal hemorrhages and anomalies of the renal pelvis. All higher doses caused maternal death. 28 Table 5. cont'd. Rats: 2,3,7,8-TCDD Mice: (i.p.) Application during the latter half (74) of gestation and in the postnatal period resulted in severe depletion of lymphocytes in the thymic cortex of the offspring. Monkeys: oral 0.05-10 mg/kg 2,4,5-T with 0.05 mg/kg TCDD No observed teratogenic effects. Mice: CF-1 0.001-3 ug/kg/day Treatment on days 6-15 by oral gavage. 3.0 ug/kg/day: fetal cleft palate and dilated renal pelvis. 1.0 ug/kg/day: cleft palate No effects at lower doses. No-effect dosage estimated at 0.1 ug/kg/day. Mice: CD-I Various dioxins 1-1000 ug/kg/day Dibenzo; 2,7-dichloro-; 2,3,7-tri(77) chloro-; 1 ,2 ,3,4-tetrachloro-; 2 ,3,7,8-tetrachloro- and octachlorodibenzo-p-dioxin administered on days 7-16 of gestation. Only 2,3,7,8 TCDD was teratogenic. Mice: NMR-1 Subcutaneous 5-40 ug/kg/day Cleft palate at all dose levels. Rats: Wistar oral 1-16 ug/kg/day Treatment during days 6-15 of gesta(78) tion caused visceral anomalies charac­ terized by edema and hemorrhage. Postnataly, the survivability, weight gain and reproductive ability of the progeny were affected. Rhesus Monkeys: 9 days by gavage total dose was 0.2 ug/kg and 1.0 ug/kg Days 20-40 All fetuses exhibited abnormalities of the soft palate. Rabbits: New Zealand White 0.1, 0.25, 0.5 and 1.0 ug/kg/day at days 6 to 15 Severe maternal and embryonal toxicity (81) at 0.25 ug/kg/day and above. Extra skeletal ribs and kidney ano­ malies at all dose levels- (76) (79) (80) 29 Table 5. cont'd. Human fetuses: collected from early abortions in Seveso, Italy. In 7 of 22 fetuses, microprecipitates observed to be regularly present beneath the trophoblastic basement membrane. No additional anomalies observed. (82) 30 teratogenic potency of 2,3,7,8-TCDD is related to its binding affinity towards the TCDD receptor (83). level in mice and rats is about 0.1 ug/kg. The observed no-effect Aside from inconclu­ sive data available from studies conducted on women exposed to 2.3.7.8-TCDD at Seveso, Italy following an industrial accident (84), which suggests an increase in birth defects, there is no direct evidence of teratogenicity induced by TCDD in human fetuses. E. Mutagenicity and Carcinogenicity A number of studies have been conducted which investigated the mutagenic potential of 2,3,7,8-TCDD. Indications of muta­ genicity were observed in tests using the African Blood Lily, which revealed chromosomal abnormalities and the inhibition of mitosis (89) . Induction of prophage in E ^ coli K-39 and strepto­ mycin independance in E^ coli Sd-4 also suggest that 2,3,7,8TCDD may be a mutagen (88) . However, the majority of the available data (Table 6) indicates 2,3,7,8-TCDD to be nonmutagenic. Additionally, Poland and Glover (103) have demon­ strated that 2,3,7,8-TCDD does not significantly bind to rat liver DNA in vivo, nor that it is metabolized to an electrophile which subsequently binds to DNA. However, the results presented in Table 7 indicate that 2.3.7.8-TCDD is extremely active in carcinogenicity bioassays. In long-term rat and mouse feeding studies, 2,3,7,8-TCDD has been shown to cause an increase in the incidence of a wide variety of neoplasms, including hepatocellular and squamous 31 cell carcinomas of the lung, hard palate/nasal turbinates or tongue; neoplastic changes in the kidney and bile duct; liver and skin carcinoma; cell adenomas. angiosarcomas and thyroid follicular The observed effects occurred following admini­ stration of extremely low doses of TCDD, suggesting that 2,3,7,8TCDD is a very powerful "complete” carcinogen. The diversity of observed neoplastic changes is uncommon to most carcinogens. In light of this diversity and the lack of evidence of DNA-binding by TCDD, it could be theorized that TCDD is actually acting as a powerful promoter. But data from two-stage mouse skin carcinogenesis bioassays does not support this conclusion. The 2,3,7,8-TCDD isomer was not shown to pro­ mote the incidence of skin tumors following initiation with 7,12-dimethylbenz(a)anthracene (99,100), (DMBA) or benzo(a)pyrene (BAP) Additionally, it wa£ shown to inhibit skin tumors when applied prior to or simultaneously with DMBA or BAP (100). However, when administered orally to partially hepatectomized rats treated with diethylnitrosamine (DEN), 2,3,7,8-TCDD acted as a promoter and increased the incidence of liver tumors (102). Regardless of the contradictory results, 2,3,7,8-TCDD clearly causes an increase in the incidence of neoplasms at very low doses. Because 2,3,7,8-TCDD also induces a range of enzyme systems and causes immunosuppressive effects at or below these levels, it is possible that the carcinogenicity arises via an indirect mechanism where TCDD is neither the material actually interacting with DNA nor the substance promoting the prolifer­ ation of cancerous cells. 32 Table 6. TCDD-induced Mutagenicity Studies. Animal and Treatment Observations Cell cultures: No significant inhibition of growth (85) in following mammalian cells:_ HeLa, Balb 3T3, virus transformed mouse fibroblasts, human fibroblasts and lymphocytes when observed for four days. No morphological changes observed by electron microscopy. 10“ 6 M TCDD Reference Rats: oral 10-20 ug/kg intraperitoneal 5-15 ug/kg Investigation of choromosomes in bone marrow: no increase in abberation rate in all cases Rats: oral 10 ug/kg. observed 0/ 24 or 72 hours following partial hepatectomy No effect on DNA synthesis in the remaining liver lobes. Salmonella tvphimurium TA 1532 and TA 1530 E. coli K-39 African Blood Lilv: tvphimurium Strains G46, TA 1530, TA 1531, TA 1532 and TA 1534. (86) (87) High mutagenic frequency in (88) Salmonella TA 1532, which re­ verts by frame-shift mutation; Negative for base-pair substitution by TA 1530; Positive for prophage induction in E. coli K-39. Induction of mutations; inhibition of mitosis. Induction of mutation in TA 1532 only. Negative in other strains (89) (90) Wistar Rats Negative in dominant lethal test. Rats: bone marrow cells Weakly positive mutagenicity. (91) Observation in absence of control. Salmonella typhimurium histidine auxotrophs. 0-20 ug TCDD per plate TA 1535, TA 100, TA 1538, TA 98 and TA 1537 strains. No strain showed mutagenicity in presence of mammalian metabolic activating systems from livers of Arochlor 1254 treated rats and hamsters or TCDD treated hamsters. (78) 33 Table 6 . cont'd. Maternal peripheral blood and fetal tissues from TCDDexposed humans at Seveso, Italy. No mutagenic effects observed. (93) 34 Table 7. 2,3,7,8-TCDD Carcinogenesis Studies. Animal and Treatment Observations Swiss Mice: TCPE + TCDD gastric intubation once weekly for one year Treatment with TCPE containing TCDD (94) as an impurity and TCDD alone at TCDD dose levels of 0.007 to 0.112 ug/kg resulted in an increase in the incidence of hepatic tumors. Mice: CD-I Two-stage skin carcinogenesis bioassay. 2 ug TCDD/mouse followed by 5 ug TPA weekly for 32 weeks. TCDD was shown to be a weak tumor initiating agent and an inactive cocarcinogen. Rats: dietary 0 .001- 0.10 ug/kg/day for 2 years. Reference (95) A two year rat feeding study at 0.01 and (20) 0.1 ug/kg/day: increase in hepato­ cellular and squamous cell carci­ nomas of the lung, hard-palate/nasal turbinates or tongue. Decreased incidence of tumors of the pituitary, uterus, mammary glands, pancreas and adrenal gland. At 0.001 ug/kg/day: no discernable effects, Rats: dietary 5 ppb in feed Death within 40 weeks. Neoplastic (96) changes observed were kidney, bile duct, liver and skin carcinoma and an angiosarcoma. Rats: dietary 1 ppt to 1 ppm in feed Increased incidence of neoplasms, (97) notably tumors of ductal epithelium and cholangiosarcomas at or above 5 ppt for 78 weeks. Suggested that TCDD is a potent promoter. Rats; oral and dermal Both TCDD and a mixture of isomers of hexachlorodibenzo-p-dioxin are weak carcinogens if administered orally, but when applied dermally they are complete carcinogens. (98) 35 Table 7 cont'd. Mice: Two-stage skin promoter assay. TCDD and 7,12-dimethylbenz (a)anthracene (DMBA) A) TCDD + DMBA: effects were addi(100) tive. B) TCDD (1 ug) 3 days prior to initia­ tion by DMBA: DMBA activity decreased by 93%. C) Dioxin applied at 0.1 ug twice weekly for 30 weeks did not act as a pro­ motor following 200 nmol DMBA. TCDD + DMBA or TCDD + BAP on mouse skin with TPA as a promoter. TCDD inhibited tumor initiation by both BAP and DMBA. Rats: Osborne-Mendel gavage 0.01, 0.05 or 0.5 ug/kg/wk in two doses for 104 weeks. Effects significant in high dose group (101) only. Male Rats: increased incidence of thyroid follicular-cell adenomas. Female Rats: increase in hepatic neo­ plastic nodules. Mice: BecSF-j^ males dosed at 0.01, 0.05 or 0.50 ug/kg wk. females at 0.04, 0.2 or 2.0 ug/kg/wk. Male and female mice: increase in hepatocellular carcinomas. Femaie mice: increase in thyroid follicular cell adenomas. Rats: orally TCDD determined to be a promoter after diethyl of liver tumors. nitrosoamine (DEN) and partial hepatectomy TCPE = tetrachlorophenyl ether TPA = t-phorbol acetate BAP = benzo(a)pyrene (101) (101) (102) 36 F. Human Effects of TCDD The effects of exposure of humans to 2,3,7,8-TCDD are not extensively documented. This is presumably due to the limited availability of TCDD-exposed human subjects and variations in estimates of exposure. Most information available today is based on studies of victims of occupational exposure and indu­ strial accidents. In many cases, the subjects underwent rou­ tine medical examinations following the appearance of acute symptoms, but long-term monitoring of chronic exposures was seldom carried out. 2,420 mg/m 2 Contamination of work areas as high as have been reported, yet few human deaths due to acute TCDD poisoning have been recorded (104). The most common symptom of human TCDD exposure is contact chloracne, along with some cases of neurological disorders and porphyria cutanea tarda. Table 8 lists the generalized symptoms of TCDD exposure in human beings. The range of toxic responses is consistent with animal data, along with subjective observations of sensory and psychological impairment. The scarcity of mor­ tality data suggests that human beings are not as sensitive to­ wards TCDD poisoning as are test animals such as the guinea pig and the rat. On this basis, it would appear that the TCDD receptor has a lower binding affinity or that human beings are capable of excreting and metabolizing TCDD more rapidly than most species tested. A number of epidemiological investigations have been con­ ducted on populations of occupationally exposed workers, which attempted to assess the relationship between TCDD and occupational 37 Table 8. Toxic Effects of TCDD Reported in Man. Dermatological: chloracne porphyria cutanea tarda hyperpigmentation and hirsutism Internal: liver damage (mild fibrosis, fatty changes, haemofuscin deposition and parenchymal-cell degeneration) raised serum hepatic enzyme levels disorders of fat metabolism cardiovascular disorders respiratory tract disorders pancreatic disorders Neurological: A) peripheral: polyneuropathie s sensorial impairments (sight, hearing, smell, taste) B) central: lassitude, weakness, loss of libido from Reggiani (104) impotence, 38 cancer. These studies have been reviewed by Tognoni (105) , Reggiani (104) and Hay (106) . Minor increases in the incidence of stomach cancers and a possible increase in soft tissue sar­ comas were observed in 3 out of 10 groups studied (104) . The observed increases did not correspond to incidents where the exposure had been the highest, but were observed in groups which were exposed to low levels of TCDD associated with the application of phenoxy acid herbicides. Regardless of the weak­ nesses inherent to epidemiological assessments, one would have to conclude that 2,3,7,8-TCDD is not as potent a carcinogen in human beings as would be expected on the basis of extrapo­ lations from laboratory animal carcinogenesis bioassays. III. A. ENVIRONMENTAL INPUTS AND FATE OF TCDD Chlorophenol Manufacturing Dioxins are not produced for any commercial purpose, but can arise as by-products of the synthesis of chlorophenols from chlorobenzene by the alkaline hydrolysis method. The formation of dioxins occurs as a condensation reaction between two ortho­ chlorinated phenol salts, as outlined in Figure 2. The 2,3,7,8- TCDD isomer can only be produced from 2,4,5-trichlorophenol. Other o-chlorophenols yield different dioxin congeners. The condensation reaction is favored at elevated reactor temperatures and pressure and becomes a major spontaneous exothermic process at reactor temperatures above 250° C. Although its formation can be minimized by maintaining a low reactor temperature, no 2,4,5-trichlorophenol produced by 39 the alkaline hydrolysis process is free from contamination by 2.3.7.8-TCDD. Levels of TCDD associated with the operation of a modern trichlorophenol manufacturing facility have been reported by Van Ness (107). Samples of aqueous effluent from the facility contained up to 100 ng/kg (ppt) TCDD and a soil sample at the site contained 559 ug/kg (ppb) TCDD. Samples of solid and liquid sludge and sump fluid contained 374, 445 and 676 ug/kg TCDD, respectively, while trichlorophenol still bot­ tom contained 111 mg/kg (ppm) of TCDD. These relatively high levels demonstrate that significant amounts of TCDD can be ass­ ociated with the production of trichlorophenol, posing especially arduous problems of occupational hygeine and waste disposal. The 2,4,5-trichlorophenol is a very important chemical inter­ mediate which has been produced in large quantities for over 40 years. During this period, standard practices of quality control and waste disposal have varied, The health hazards associated with 2,3,7,8-TCDD have not been fully realized until very recently. In fact, during the last major TCDD accident, which occurred in Seveso, Italy in 1976, the officers of the plant were reportedly unaware of the hazards of dioxins until well after the explosion (108). Recently, severe contamination by 2.3.7.8-TCDD has been reported in Missouri as a result of dis­ posal of chlorophenol production wastes by application to roads as a means of dust control (109). It is quite likely that a number of similar instances have taken place unreported and are contibuting to the presence of environmental residues of 2.3.7.8-TCDD. 40 B. TCDD release during industrial accidents There have been a number of serious accidents which oc­ curred during the manufacture of 2,4,5-trichlorophenol from chlorinated benzene in which a chlorophenol reactor overheated, proceeded to form large amounts of dioxins and exploded (108). A large number of serious cases of occupational exposure to 2,3,7,8-TCDD have also been reported, presumably the result of handling TCDD-contaminated chlorophenols;. products made from 2.4.5-trichlorophenol and during routine operations of plant maintenance and waste disposal. The most recent accident took place in Seveso, Italy in July of 1976 when a chemical reactor in a factory producing 2.4.5-trichlorophenol for use in hexachlorophene synthesis overheated, went exothermic and blew its contents out a safety valve which was vented to the outside of the plant. The ex­ ploded contents of the reactor formed a cloud which was carried away from the factory, heavily contaminating a 700 meter by 5 km section of the city with 2,3,7,8-TCDD. inhabitants A week later, the of the town were told what had happened, and two weeks later, persons living in the area most heavily contami­ nated were evacuated from their homes. The amount of TCDD re­ leased has not been determined, but estimates range between 300 grams and 130 kg (110). There was an estimated total of 49,500 kg of soil into which TCDD had penetrated and lodged, and many homes and schools were severely contaminated. Removal was judged impossible and burning was ruled out due to the 41 thermal stability of TCDD and the possibility of recontamination. Dogs, cats, rodents and birds present in the effected area died within a short time but interestingly, the only human effects reported were a large number of serious cases of con­ tact chloracne. To date, there have been no fatalities recorded as a result of this incident. Women who were pregnant and ex­ posed were urged to undergo abortions, while those of childbearing age were warned not to attempt to conceive children. There have been two children with deformities born to exposed women. The first abnormality observed was a slight intestinal obstruction in an infant born to a woman who had reportedly eaten contamin­ ated garden produce; while a woman living in an area not offi­ cially considered contaminated gave birth to a child with a minor genital malformation (11). Both children were born at a date which indicated that they were beyond the teratogen-sensitive period of organogenesis at the time of the exposure, and it has not been determined whether they represent an increase over normally expected background levels. Currently, a one square kilometer area of Seveso has been sealed off from access and will probably remain unusable for at least 15 years. Much of the decontamination of other exposed parts of the city has been carried out by removal of ground cover and by the ultraviolet photolysis of the interiors of homes. 42 C. Product Contamination 1. Phenoxy acid herbicides Most products using 2,4,5-trichlorophenol as an inter­ mediate have been shown to contain varying amounts of 2 ,3,7,8TCDD as an impurity. The germicide hexachlorophene; an herbi­ cide of major importance, 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) and its propionic acid analog, Silvex, are all syn­ thesized from 2,4,5-trichlorophenol. Prior to 1971, these products commonly contained up to 50 mg/kg (ppm) of 2,3,7,8TCDD. However, following the recognition of the toxic potential of TCDD, steps were taken to reduce the contamination by closely monitoring conditions during the production of the trichlorophenol intermediate. An industry standard for TCDD in 2,4,5-T has been set at 0.1 mg/kg, which is the typical TCDD level found in much of the 2,4,5-T produced today. Nevertheless, the United Sates Environmental Protection Agency has suspended the registra­ tion of 2,4,5-T and Silvex in the U.S. A number of studies have been conducted to determine whether residues of TCDD could be found in areas sprayed with 2.4.5-T (112-117). However, due to the low levels of application of TCDD and the analytical sensitivity required to detect the TCDD residues, most investigators reported negative findings. A study conducted by Woolson (116) analyzed soil core samples taken from a sandy area in Florida which had been sprayed with 2.4.5-T at application rates up to 11 g/m 2 (947 lbs/acre). Application was conducted during the period 1962-1969, when 43 TCDD levels present in 2,4,5-T ranged between 30-50 mg/kg. This study detected no traces of TCDD in the soil at a minimum detectable level of 1 ug/kg. Another series of experiments was conducted to determine whether residues of TCDD were present in environments routinely treated with 2,4,5-T (112-114). studies were conducted with purified 2,4,5-T. The Samples of fish, water, mud, human milk, cow milk and beef fat were analyzed by an analytical procedure sensitive to part per trillion (ng/kg) levels of TCDD. None of the samples analyzed were found to contain TCDD, with the exception of samples of beef fat which were positive at 3 ng/kg. A study conducted by Newton (115) demonstrated that 2,4,5-T containing 20 ug/kg or less applied to a forest ecosystem did not significantly effect the health of the forest herbivore, the mountain beaver. There were no residues of TCDD detected in the livers of trapped beaver at a detection limit of 5 to 10 ng/kg. However, other studies have shown that cattle grazing in rangeland sprayed with 2,4,5-T con­ tained residues of TCDD at levels ranging from 5 to 143 ng/kg (117). The inconsistencies in the residue data most likely reflect the variability of residues of 2,3,7,8-TCDD in the 2,4,5-T used for application. 2. Agent Orange During the Vietnam War, a 1:1 mixture of 2,4,5-T and 2,4,-D (2,4,-dichlorophenoxyacetic acid) known as "Agent Orange" was sprayed over a very large portion of Vietnam by the military to defoliate the terrain. of the mangrove forests, More than 10% of the inland forests, 36% 3% of the cultivated and 5% of "other" 44 land was sprayed one or more times (118). A total estimated volume of 45 million liters of Agent Orange, which contained 2,3,7,8-TCDD at levels as high as 47 mg/1 was applied. There have been many claims made by veterans of the war in Vietnam as well as by the Vietnamese natives that Agent Orange was the cause of various cancers and birth defects. At the present time, an epidemiological survey is being conducted on veterans and their families to determine whether such a relationship exists. A study conducted by Baughman (120) analyzed fish and crustaceans collected from Vietnam for 2,3,7,8-TCDD. TCDD was detected in all of the samples at levels ranging from 18 to 810 ng/kg. The highest concentrations of TCDD were measured in catfish taken from the Dong Nai river in the interior, where spraying was reportedly the heaviest. A study conducted by Young (121) on a 2.6 square kilometer military test site which had been sprayed with a total of 73,000 kg of 2,4,5-T and 76,000 kg of 2,4-D detected 10 ng/kg to 710 ng/kg 2,3,7,8-TCDD in the top 15 cm of soil at the site. In some instances, ten years had elapsed since the last aerial application. Liver tissues of rodents inhabiting the test site contained 210 to 1,300 ng/kg TCDD, but no gross or histological evidence of carcinogenesis, teratogenesis or toxicity was observed in 112 adults and 87 fetuses. Concentrations of TCDD (12 ng/kg) were detected in two species of fish in a stream running through the site. Levels of 2,3,7,8-TCDD in skin, muscle, gonads and gut of a spotted sunfish from a test pond at the site were 4, 4, 18 and 85 ng/kg, respectively. 45 3. Pentachlorophenol Significant levels of dioxins have also been detected in technical grade pentachlorophenol (122-125) . A study conducted in our laboratory on an industry composite sample of technical pentachlorophenol (PCP) measured over 1,000 mg/kg each of hexa- and heptachlorinated dioxins and 900 mg/kg OCDD (126). The 2,3,7,8 isomer of TCDD was not detected at a limit of detection of 0.1 mg/kg, but other tetrachlorinated isomers were detected at 1 mg/kg. Technical grade PCP has been demonstrated to induce the type of toxicity characteristic to dioxin poisoning while puri­ fied PCP has not. PCP is widely used in the preservation of wood, and is probably one of the most versatile pesticides in use today. Dioxins have been shown to be associated with the lumber and pulp industry, where PCP is used as a slimicide and preservative (127). Shavings, sawdust and wastes are routinely burned, resulting in the formation and release of dioxins. Due to its wide use and its relatively high dioxin content and po­ tential for formation during burning, PCP should be considered a dangerous material. A study by Firestone (128) investigated the accumulation of dioxins in cows fed technical grade P C P . Cows were fed PCP containing 20, 280, and 700 mg/kg hexa, hepta and octachlorodioxin at PCP dose levels of 20 mg/kg/day for 10 days and 10 mg/kg/day for the next 60 days. Composite milk fat contained 85 ug/kg total dioxins at the end of the treatment period, whereas total dioxins in the blood were 0.25 ug/kg. 46 D. Formation of Dioxins During Combustion In 1977, Olie (129) conducted a survey of the organo- chlorine residues present in fly ash and made an observation which suggested that chlorinated dioxins are produced during combustion. He detected varying amounts of chlorinated ben­ zenes, phenols, dioxins and dibenzofurans in samples of fly ash and flue gases from three municipal incinerators in Holland. These materials could not be accounted for on the basis of their presence in the waste prior to incineration. It was theorized that the dioxins were formed as a result of the con­ densation of chlorinated phenols produced de novo by reaction of organic constituents of the waste such as polyethylene with inorganic chlorides under the pyrolytic conditions in the incinerator. Since that time, a number of investigators have consistently detected dioxins in fly ash and exhaust gases from municipal and chemical waste incinerators in Europe, the United States, Canada and Japan (130-139). are listed in Table 9. The results of some of these studies The dioxins present at the highest con­ centrations are the higher chlorinated, tetra through octachlorinated congeners. most abundant. Of these, the hexachlorinated isomers are the Reported total dioxin levels in fly ash range from 200 to over 5,000 ug/kg, with TCDD levels ranging from 2 to over 300 ug/kg (138). Few studies have specifically measured the 2,3,7,8-TCDD isomer. This is mostly due to the difficulty of isomer-specific dioxin analysis. Most recently, Tierman (140) identified 20 of 47 the 22 tetrachlorinated isomers in a sample of municipal waste fly ash from the U.S. The 2,3,7,8-TCDD isomer was present in the greatest concentration (2.1 ug/kg), representing 15% of the total TCDD content of the fly ash. Researchers at the Dow Chemi­ cal Company in Midland, Michigan estimate flue gas from municipal incinerators to contain an average of 2 ng 2,3,7,8-TCDD per Nor3 mal cubic meter (ng/Nm ) (141). On the basis of a ton of waste 3 producing 30 kg of fly ash and 7000 Nm of flue gas; a ton of waste should yield an estimated 63 and 14 ug of 2,3,7,8-TCDD in fly ash and flue gas, respectively. A comprehensive survey of PCDDs and PCDFs and related compounds present in incinerator effluents was conducted by Olie (138). Average and peak values for PCDDs, PCDFs, chlori­ nated benzenes and phenols were obtained by sampling the major municipal incinerators in Holland over a two year period. On the basis of data collected, they estimated the total amount of PCDDs and PCDFs produced by incineration of municipal waste on a yearly basis to be 145 kg of total dioxins and 97 kg total chlorinated dibenzofurans (Table 9 h ) . They further estimated the toxic potential of the PCDDs and PCDFs produced to be equi­ valent to 10 kg of 2,3,7,8-TCDD/yr in fly ash and 2 kg/yr in flue gas. These figures were arrived at by assigning a toxicity factor of 1 to 2,3,7,8-TCDD and a factor of 0.1 to the other compounds with at least three chlorine atoms in the lateral positions and at least one hydrogen at the peri positions. When estimated in this manner, 2,3,7,8-TCDD represents only about 2% of the toxic potential of fly ash and incinerator 48 Table 9. Dioxin Levels Associated with Incineration of Municipal Wastes. a. Municipal and industrial waste: TCDD Municipal waste: (Ref. 132) PgCDD HgCDD H^CDD OCDD (ug/kg) ash collected from oven at 900°C. N.D. N.D. N.D. N.D. N.D ash collected from stack at 260°C. 2 8 30 60 120 Industrial waste incinerator: 100 160 180 130 140 N.D. = not detectable b. Municipal incinerator fly ash collected from electrostatic precipitator (E.S.P.): 3 samples. (Ref. 133) TCDD:6 , 1.1 and 2.7 ug/kg OCDD: c. 150, 30 and 0.7 ug/kg Municipal waste incinerator fly ash (E.S.P.): 5 samples: TCDD: 12, 10, 9, 5,and 2.5 ug/kg (Ref. 134) 49 Table 9. cont'd. d. Municipal waste: results of 80 analyses from 25 different incinerators (ug/kg) (Ref.135) TCDD P 5CDD HgCDD H7CDD OCDD low value! (E.S.P.) 5 31 80 190 266 median value (E.S.P.) 54 182 326 288 106 high value (E.S.P.) 110 488 1200 902 110 Stack particulates 100 800 1370 1370 310 e. Variation of PCDD content of fly ash (E.S.P.) from one incinerator over a 4-week period (ug/kg) (Ref.134.) TCDD P-CDD $ ... HgCDD H 7CDD week one 81 252 406 321 82 week two 56 173 388 385 124 week three 54 182 326 288 106 week four 41 159 381 458 226 f. OCDD Variation of PCDD content of flue gases collected (Ref. 136) over a 9-month period from one municipal incinerator. 3 nanograms/Normal cubic centimeter (ng/N cm ) TCDD ---- P..CDD —D H^CDD H„CDD OCDD —O — I ---- high value 1127 2089 3805 2884 631 mean value 128 260 366 286 126 low value 7 19 15 13 19 50 Table 9 cont 'd. g- 1: 2: 3: 4: 5: 6: (Ref. 137) Dioxin levels measured at 6 different municipal incinerators: TCDD P 5CDD HgCDD H 7CDD OCDD stack particulates * 1.1 2.7 11.5 1.0 8.0 flue gjas * 19.6 27.9 178.2 159.6 63.9 E.S.P. fly ash (ug/kg) 0.25 1.7 294.0 8.9 295.0 stack particulates 172.2 172.3 12,015 575.0 7,312 flue gas 17.0 107.0 26,620 828.0 1,179 E.S.P. fly ash (ug/kg) N.D. 0.9 1.8 3.1 1.5 stack particulates 0.037 0.3 6.7 0.2 1.7 flue gas 19.0 40.0 65,420 124.0 776.0 E.S.P. fly ash (ug/kg) 46.4 65.4 2,496 87.9 841.5 stack particulates 10.9 2.8 0.5 3.2 39.0 flue gas 60.0 33.0 1,390 167.0 2,703 E.S.P. fly ash (ug/kg) 0.7 0.05 0.02 0.01 0.1 stack particulates 0.34 2.4 196.0 9.9 173.0 flue gas 9.6 21.0 328.0 46.0 244.0 E.S.P. fly ash (ug/kg) N.D. N.D. N.D. 0.001 5.9 stack particulates N.D. 0.01 0.28 N.D. 0.5 flue gas 19.0 11.0 480.0 6.0 71.0 E.S.P. fly ash (ug/kg) * Stack particulates and flue gas levels reported as nanograms 51 Table 9. cont'd. h. Average concentration and estimated yearly output of PCDD and PCDF in fly ash and. flue gas of the major municipal incinerators in the Netherlands (reported as the total of all incinerators sampled.) (Ref. 138) flue gas average - total amount conc.(ng/Nm ) per year (kg) TCDD 93 5.6 57 o • 00 fly ash average * total amount conc. (ug/kg) per year (kg) P,.CDD 254 15.2 244 3.4 H^CDD o 604 36.2 440 6.2 H?CDD 760 45.6 347 4.9 OCDD 345 20.7 452 6.3 total 2056 123.3 1540 21.6 173 10.4 161 2.3 PgCDF 312 18.7 272 3.8 H.CDF 6 459 27.5 528 7.4 H?CDF 314 18.8 293 4.1 OCDF 51 3.1 68 1.0 total 1309 78.5 1322 18.6 (Furans) TCDF 52 gas. An early life stage fish bioassay using extracts from municipal fly ash supports this estimate (142). Fly ash extracts were determined to be more than 10 times as toxic as would be expected on the basis of 2,3,7,8-TCDD present in the extract. Therefore, the majority of the toxic potential of the dioxin type associated with fly ash is the result of the less toxic dioxin and dibenzofuran congeners which are present in fly ash at much higher concentrations than 2,3,7,8-TCDD. Olie (138) also noted that the amounts of dioxins pro­ duced and the ratios of the numerical isomers vary with temper­ ature, the point in the incinerator where they are collected and among the various incinerators. No dioxins were detected in waste slag, while the concentrations and congener ratios associated with fly ash collected from electrostatic precipita­ tors are different from those escaping with flue gas. Flue gases contain less PCDD than either type of particulate. This indicates that the formation of PCDD is a dynamic process taking place on the surface of particulates, which may be acting as catalytic sites. Eiceman (143) investigated the behavior of 1,2,3,4-TCDD adsorbed onto fly ash by simulating emission conditions in a municipal incinerator. He demonstrated gas-stream conversion of adsorbed 1,2,3,4-TCDD to P^CD, HgCDD, H^CDD and OCDD in the presence of as little as 1% HC1 gas in air. The value of 1% HC1 is within the range of HCl levels present in modern munici­ pal incinerators. were monitored. Reaction temperatures from 50°C to 250°C The highest conversion was observed at 150°C. Below this temperature, the heat of reaction for chlorination 53 of dioxin (an endothermic process, A h =+71 kJ/mole/chlorine) was not efficiently achieved. The reduction in conversion at the higher temperature was believed to be due to stronger adsorption of the dioxin molecule to the particle surface, which hindered interaction with the gas stream. These results partially explain the observed variations in isomer ratios and the predominance of the higher chlorinated congeners. There are a number of possible explanations for the presence of dioxins in combusted materials. The most straightforward is the assumption that dioxins are already present in the materials being combusted. They may be present as a result of dioxin contamination in herbicide formulations, in polychlorinated biphenyls or as components of halogenated phenols such as pentachlorophenol. However, this would require a much higher level of initial contamination than can be explained by a mass balance. Analyses of compost (144) and municipal waste prior to incineration (145) have not revealed the presence of any dioxins. An explanation which reflects a good deal of experimental evidence is that the dioxins arise from chlorinated precursors of anthropogenic origin present in the waste being burned. Laboratory scale combustion studies have shown that 2,4,5-T (146), chlorophenols (147-149), PCBs benzenes (150,151), chlorinated (152), chlorinated aliphatics (155,156) and chlorinated polymers, such as polyvinyl chloride (153) can yield chlorinated dioxins and dibenzofurans or their precursors. Mechansisms of their formation are presented in a review by Choudry (154). Figure 4 outlines the reactions which various organic materials P C O fs or COAt Alkyl PVC PCDD s PCDD* LCNINS PCB's -HCl C H C I3 DOT CO, PETROLEUM COKE Figure 4. Routes of Formation of Chlorinated Dioxins During Combustion. (From Choudry et al. Ref. 154) 55 undergo to form PCDD and PCDF during combustion. Chlorinated phenols yield PCDD via the previously outlined condensation reaction (Fig. 2). The herbicide 2,4,5-T and similar chlorinated phenoxy acid compounds should also form dioxins by condensation following cleavage of the ether linkage. Ortho chlorinated phenols are the ultimate precursors required for formation of PCDD from chlorinated waste constituents. Any material which can react to produce an ortho chlorinated phenol during combustion will yield dioxins, which can be further chlorinated via gas-phase reactions with HC1 or C I 2 in the incinerator stack at the appropriate temperature and level. Chlorobenzene is a key intermediate in the thermal forma­ tion of chlorophenols. It reacts via a free radical mechanism in which a hydrogen atom on chlorobenzene is abstracted by a chlorine or hydrogen free radical to yield a chlorobenzene free radical and HC1 or Hj* In the presence of oxygen, the chlorobenzene free radical reacts to produce the chlorophenol (152) . Hexachlorobenzene (HCB) and pentachlorobenzene are usually the major chloroaromatic constituents present in combustion train exhaust gases containing chlorinated dioxins. It is believed that hexachlorobenzene undergoes reductive dechlori­ nation to pentachlorobenzene which can further react via free radical mechanisms to form chlorophenols and ultimately PCDD and PCDF. The source of the HCB can be directly anthropogenic, it can arise via gas phase chlorination of lower chlorinated benzenes or it can be formed from chlorinated aliphatics. inated aliphatics which satisfy the relation: Cx C l ^ j Chlor­ l 39 1ST SC/PG: 2.3.6 TP ICHI.OROPHENOL PvPOLVZATE IP 1 12. 2,3,6-trichlorophenate pyrolysis products »* SELECTED JON CHROHATOCRAII tt S.3,*/2,3,5 TRICHLOROPHENOL PYROLYZATE FPII 1ST SC'PC: 6333 1 1.0 UL H273. 17 7d Products of 2,3,4- plus 2,3,5-trichlorophenate pyrolysis 107 rr SELECTED ION CHRONATOGRGN «* 3,3. V 2 . 3 . 6 TRICHLOROPHENOL PYROLYZATE 1.0 UL H6S96. FRfl 6094 1ST SC/PC: 1 *• .SO V. 1.00 12 6 17 18 134.0 11? , . 333.0 L i __A A 330.0 i '■* 7e . 1 I S .... 1 IS L1 A A I 17 A A 10 L »0 PC* 31 Products of 2,3,4- plus 2 ,3 ,6-trichlorophenate pyrolysis XX SELECTED ION CHROnftTOCRAn l» 2,3. 4/2.4, S TRICHLOROPHENOL PYROLYZATE 1.0 UL FRN 6036 1ST SC/PC* I K* .50 V« I .00 16 18 14 322. d 7f a9 OI9 A-iV T -Aa .1 7 _L£_ 1r 0 T = 0 T1 3 Products of 2,3,4- plus 2,4,5-trichlorophenate pyrolysis 10S a,3,5/2,3,S •t.t SELECTED ION CHKOI’IATCCRMI *i TRICHLOROPHENOL PYROLYZATE 1.0 UL ia 12 7g 1ST SC/PG X* 20. 19. Products of 2,3,5- plus 2,3,6-trichlorophenate pyrolysis PPM €093 t* SELECTED ION CHROMATOGRAM xx 1ST SC'PCi Z,3.6/2.4,5 TRICHLOROPHENOL PYROLYZATE l.i UL X* 1 .5 0 V ' 1 .0 0 16 14 17 DeS.a. I 14 7h J£- I 17 WJ1 'I :a _£2_ pi Products of 2,3,6- plus 2 ,4 ,5 -trichlorophenate pyrolysis 109 ' F3N 6093 1ST SC'PC: 1 X.50 V* 1.00 St SELECTED ION CHROnftTOGSPH 2 ,3 ,4 .5/S,4 CHLOROPHENOL PvROLYZPTE 1.0 UL I 13 322.0 T“ 1±. 7i _1£_ MS JLZ- 1R I _La_ _2fl. _ai_ Products of 2,3,4- plus 2,4-dichlorophenate pyrolysis » * SELECTED ION CHROnATOCRAfl *» 2,3.4,6Sa,e CHLOROPHENOL PYROLVZATE PHN 6101 1ST SC'PGt i 1.0 UL H1917. 7j Pyrolysis products of 2,3,4- plus 2 ,6-dichlorophenate 110 xt SELECTED ION CHROCIATOCRAH «* 2,3,S,6/2.« CHLOROPHENOL PVROLVZATE 1.0 UL 1516 17 12. IS. 7k Pyrolysis products of 2,3,5- plus 2,4-dichlorophenate X* SELECTED ION CHROMATOCRAN «< 2.3. S.6/2.S CHLOROPHENOL PVROLVZATE UL 1.0 FRN S102 1ST SC/PC: 1 X* .50 V» 1.00 7 1 22 12 322. C 14 71 15 Ifi 6 I? 1 17 1 ia 19 1 aa « ai Pyrolysis products of 2,3,5- plus 2,6-dichlorophenate Ill rf SELECTED ION CHRONRTCCRArt «7 2, 7 , 4 . 5/2. 4 *2.5 NC 2 , 3 . 7 , 8 t c » p MIX FPR RESOLUTION CHECK H135. H* 230. FRN 6103 1ST SC/PGi 1 X* 1.00 V 1.00 14 13 i I I I _ L 4 _ -----------------1 5 . 7m _ I . Ai » 8Ji 1 IS 1 1? Resolution of 2/3/ 1 , 8-TCDD from nearest eluting TCDD isomers. St SELECTED ION CHRCn«TOCRfln tx 2.3.7.a TCDD-1.0 NC 7-25-32 2.0 UL OF O.SNG/UL H- 7n 19 GC/MS chromatogram of 2,3,7,8-TCDD 112 E. Quality Control/Quality Assurance 1. Method validation Each step in the analytical procedure was individually validated and maximized for recovery of 2,3,7,8-TCDD by use of a radioactive tracer. A 2.0 ng aliquot of (specific activity: 126 mCi/mmole) same manner as the sample. 14 C-2,3,7,8-TCDD in hexane was treated in the Quantification of recovery was accomplished by liquid scintillation counting. The extracts were reduced to a volume of approximately 3 ml, transferred into 15 ml of scintillation cocktail (Omnifluor; New England Nuclear), counted in a Searle, Isocap 300 liquid scintillation counter and quantified with respect to an external standard quench-corrected standard curve. The optimal observed recoveries are listed in Table 14. Most steps in the procedure exhibited a greater than 90% recovery. However, the product of losses encountered at each step results in an expected overall recovery of between 10 and 60%. Although they may be a little low, the recoveries are satisfactory. They do not need to be consistent because the internal standard is used to correct for losses. However, a 10% recovery raises the theoretical lower limit of detection by a factor of 10, thereby lowering the sensitivity of the technique. The highest losses were encountered during the HPLC and GC/MS transfer steps. The losses were caused by inefficient reconstitution and transfer of the sample extract. These steps require the entire sample extract to be dissolved and transferred with a minimum volume of solvent. Final GC/MS quantification 113 requires the total residue to be extracted into and transferred in only 2 ul of isooctane. Although any portion of the sample remaining in the vial can be used for subsequent quantification, the greatest sensitivity is achieved with an efficient transfer. The recoveries listed in Table 14 are a range of maximum recoveries observed after the method was validated and maximized for recovery. There were many instances of extremely low recovr eries encountered during initial method validation which identified potential quality control problems. Use of the 14 C-2,3,7,8- TCDD tracer was an integral part of the quality control program utilized in this work. It was especially useful because it could be applied at levels (1 to 2 ng) similar to those anticipated in the actual samples, making it an excellent tracer of low-level adsorptive losses; yet it could be easily detected and quantified using standard liquid scintillation technique. A major quality control problem identified by use of the radioactive tracer was variation in adsorbant activity. The activity of adsorbents, especially the basic alumina, varied between batches resulting in elution of the dioxin fraction with the CC1 4 :hexane fraction instead of eluting later with methylene chloride. An adsorbant quality control program was established to minimize such error. The activity of each batch of adsorbants was evaluated following purification and activation. 114 Table 14 Optimum Working Recoveries Determined Using 14C-2,3,7,8-7000 Treatment % Recovery A. Acid digestion/hexane extraction 95.0 B. H 2SO4 wash 91.8 C. Rotary evaporation 95.6 - 98.6 D. ^SO^-Silica 78.2 - 94.8 E. NaOH-Silica 84.8 - 89.0 F. Rotary evaporation 95.6 - 98.6 G. AgNO3-Silica 99.8 H. Activated Alumina 95.5 I. N-Evaporation 92.1 - 96.5 J. HPLC Purification 52.8 - 94.1 K. Hexane extraction 91.0 - 99.8 L. N-Evaporation 92.1 - 96.5 M. GC/MS transfer 52.8 - 94.1 Total procedure recoveries were calculated as the product of the recoveries measured for each step. % Recovery = (A x B x C ...K x L x M) Range of % Recovery = 10.8 - 56.1% 115 Significant losses were also observed during sample evaporation by Femtogas nitrogen. It was determined that losses could be minimized by using a very gentle stream of nitrogen and not heating the extracts to above 3 0 °C, thereby preventing volatization or aerosol formation. The extracts were removed from the gas stream just before going dry. If they were to be stored for any length of time, they were reconstituted into 5 ml of hexane and stored at - 1 5 °C. Another step at which major losses were encountered was during the loading of the HPLC injection loop with the sample extracts for HPLC purification. Due to non-laminar flow conditions in the sample loop, attempts to load more than 32 ul of extract into the 50 ul sample loop resulted in the establish­ ment of a leading volume of extract which flowed out of the sample loop before the entire sample was loaded. This loss was minimized by maintaining an injection volume of less than 32 ul and by filling the loop with a slow, even motion. A 100 ul sample loop was later installed, further improving the efficiency of this step. 2. Reagent purity In order to prevent the establishment of false positive values and reduce background noise, the solvents, adsorbents and gases used were of the highest possible purity. Nitrogen for evaporation and the clean-up adsorbents were purified by the methods reported in the previous section. Extracts of the 116 adsorbents or solvents were concentrated 1000:1 and evaluated by UV-HPLC. Entire batches of adsorbents or solvents were discarded if their extracts showed any UV-HPLC activity in the TCDD region. The extracts could not be evaluated by GC/MS due to limited availability of a functioning mass spectrometer. No wipe tests or blanks could be evaluated during the sample work-up period. All glassware used was washed in hot, soapy water (Alconox), rinsed in distilled water, and then wash-bottle rinsed with distilled, deionized water, acetone and then glass-distilled hexane. The glassware was then placed in an oven and baked at 400°C for a minimum of 12 hours. There was no quality control procedure which would have alerted the investigators to low-level contamination by 2,3,7,8TCDD. This study relied solely on a set of procedural blanks for evaluating sample contamination. However, these were not analyzed until all of the fish samples had been processed and GC/MS conditions were optimized for detection of 2,3,7,8-TCDD. Fish samples were processed in batches of 20, with the inclu­ sion of two procedural blanks per batch. Following GC/MS quantitation, all of the recoverable blanks were determined to be free of dioxin contamination at various limits of detection. A list of the procedural blanks and their lower limits of de- . • tection are presented in Table 15. 117 Table 15 Procedural Blanks I.D. number 2,3,7,8-TCDD (ng/kg) Limit of Detection (ng/kg) 1 N.D. 28 2 N.D. 9 3 N.D. 4 4 N.D. 10 5 N.D. 3 6 N.D. 5 7 N.D. 8 8 N.D. 20 9 N.D. 7 10 N.D. 53 11 N.D. 44 12 N.D. 36 13 N.D. 20 N.D. = no 2,3,7,8-TCDD detected at corresponding limit of detection. Blanks contained no fish, but values calculated on the basis of a 20 g sample of fish with a minimum detectable mass of 20 pg of TCDD. 118 3. Maximization of selectivity towards 2,3,7,8-TCDD The specificity of the technique for 2,3,7,8-TCDD was assured by isolating the 22 TCDD isomers and maximizing their capillary GC resolution. The previous section demonstrates that resolution of 2,3,7,8-TCDD from the other TCDDs was achieved. The satisfactory resolution of 2,3,7,8-TCDD, combined with use of the 13 C-2,3,7,8-TCDD internal standard and the simultaneous monitoring of the 319.9, 321.9 and 323.9 mass fragment cluster of TCDD assures that a peak with a GC retention time equivalent to 2,3,7,8-TCDD is indeed 2,3,7,8-TCDD. This protocol gave reasonable assurance that the compound being quantified was in­ deed 2,3,7,8-TCDD. A more positive confirmation could only have been achieved with a high resolution mass spectrometer. F. Results of 2,3,7,8-TCDD Determinations Listed in Table 17 are the levels of 2,3,7,8-TCDD detected in samples of fish collected from, various rivers in Michigan and from the Great Lakes. A sample log listing the species, weights, lengths and age of the samples is contained in Table 16. A map of the lower peninsula of Michigan identifying the location of each sample site is presented in Figure 8 . The lower limit of detection of each sample was determined on the basis of a minimum detectable mass* of 20 picograms of 2,3,7,8-TCDD corrected relative to the weight of sample aliquot of fish and the mass of the internal standard detected in the injection aliquot. A 20 g sample of fish with a 100% recovery 119 of internal standard would have a lower limit of detection of 1 ng/kg, or 1 part per trillion of 2,3,7,8-TCDD. Due to variations in sample recoveries, the limit of de­ tection of each sample is unique and should be considered when evaluating the significance of a sample which contained no de­ tectable residue of 2,3,7,8-TCDD. In a number of cases, samples had detectable residues of 2,3,7,8-TCDD at levels much lower than the limit of detection for samples in which no TCDD was detected. In addition to the fish residue survey, two short residue experiments were conducted, the results of which are listed in tables 18 and 19. The first experiment involved the analysis of liver samples from mink maintained on a diet of Saginaw Bay carp. The liver samples were analyzed in the same manner as the fish samples, with the exception that a sample size of about 12 g was used. The 12 g sample represented the entire mink liver. The second short experiment involved the investigation of the effect of cooking on levels of TCDD present in the fish sample prior to treatment. Patties formed from the ground fish samples were cooked by broiling and analyzed for the presence of 2,3,7,8-TCDD. The residues were calculated on the basis of the final weight of the cooked fish and compared to the levels of 2,3,7,8-TCDD present in the fish prior to broiling. 120 Table 16 Fish Sample Log SNumber Species Length(cm) Weight(kg) Age(yr) 1 White Sucker 40.5 0.759 5 2 Carp 51.0 1.979 5 3 Redhorse. Sucker 46.3 0.80 5 4 Carp 68.5 4.84 10 5 Carp 50.2 1.90 7 6 Carp 33.0 0.797 4 7 Carp 45.7 1.66 6 8 Carp 51.8 1.84 7 9 Carp 50.6 1.88 6 10 Carp 44.0 1.03 5 11 Carp 26.7 0.34 4 12 White Sucker 39.5 0.80 3 13 White Sucker 46.0 0.98 3 14 Carp 41.2 1.03 5 15 Carp 53.3 1.42 7 16 Carp 50.2 1.28 6 17 Carp 31.7 0.52 9 18 Carp 51.0 2.08 10 19 Carp 57.0 2.42 7 20 Carp 50.4 1.54 8 21 White Sucker 43.7 0.83 2 22 Carp 68.6 5.10 8 23 Carp 55.9 3.06 6 24 Carp 33.0 0.85 7 25 Carp 45.4 1.82 9 26 Carp 52.0 1.98 8 27 Carp 47.0 1.34 7 28 Carp 45.1 1.75 8 29 White Sucker 41.0 0.89 5 30 White Sucker 43.1 0.96 4 121 Table 16 (con't) Sample Number Species Length (cm) Weight.(kg) Age(yr) 31 Carp 51.0 1.91 4 32 White Sucker 39.7 0.65 3 33 Carp 52.1 1.76 6 34 Carp 33.6 1.07 8 35 Carp 88.9 5.30 8 36 White Sucker 43.2 0.89 5 37 Redhorse Sucker 32.7 0.43 5 38 Carp 45.7 1.37 10 39 Carp 43.6 1.45 6 40 White Sucker 34.5 0.45 3 41 White Sucker 31.7 0.32 2 42 White Sucker 33.6 0.41 2 43 Carp 38.1 0.77 5 44 Channel Catfish 44.4 1.23 _ 45 Carp 53.5 2.22 8 46 Carp 52.0 2.00 8 47 White Sucker 43.1 0.96 4 48 Carp 49.4 1.73 6 49 White Sucker 43.6 1.02 4 50 Carp 49.4 1.73 4 51 Carp 50.0 1.78 5 52 Redhorse Sucker 43.3 1.00 5 53 Carp 3 8 ■*1 0.63 7 54 Carp 53.5 2.40 4 55 Carp 60.5 2.65 6 .(56 White Sucker 48.3 0.73 4 57 White Sucker 51.2 1.565 4 58 Carp 44.0 1.31 3 59 Carp 41.6 1.12 3 60 Carp 38.4 1.21 3 61 Carp 44.6 1.42 4 62 Carp 35.5 0.78 3 122 Table 17 Results of Fish Analyzed for 2,3,7,8-TCDD 125 (34) 1 1 St. Joseph River at Benton Twp Park 2 2 Kalamazoo River at Saugatuck 47 2 Kalamazoo River at Saugatuck n.d. (33) 48 2 Kalamazoo River at Saugatuck n.d. (21) 49 2 Kalamazoo River at Saugatuck n.d. (76) 50 2 Kalamazoo River at Saugatuck n.d. (5) 51 2 Kalamazoo River at Saugatuck n.d. (115) 5 3 Grand River at Spring Lake 324 (152) 52 3 Grand River at Spring Lake n.d. (20) 53 3 Grand River at Spring Lake n.d. (38) 54 3 Grand River at Spring Lake n.d. (173) 6 4 Grand River at Grand Ledge 298 (158) 3 5 Muskegon River at Cobb Power Plant 123 (14) 55 5 Muskegon River at Cobb Power Plant n.d. 4 6 Muskegon River at Bridgeton 237 (141) 56 7 Muskegon River at US-131 n.d. (56) 57 8 White River at White Lake n.d. (48) 58 8 White River at White Lake n.d. (34) 29 9 Black River at Tower, MI. n.d. (88) 30 9 Black River at Tower, MI. n.d. (13) 31 10 AuSable River above Oscoda, MI. n.d. (40) 32 11 AuSable River at Mouth n.d. (35) 13 12 Tittabawassee R. 8 km upstream to Dow 287 (84) 14 12 Tittabawassee R. 8 km upstream to Dow 20 (14) 36 12 Tittabawassee R. 8 km upstream to Dow n.d. 10 13 Chippewa R. 16 km upstream to Dow 136 (36) 11 14 Pine River, below St. Louis 322 (34) 12 14 Pine River, below St. Louis 85 (67) • 214 (50) (51) (63) 123 Table 17 cont'd 2,3,7,8-TCDD ng/kg Map I.D. Location 15 15 Tittabawassee, below Dow dam 17 (12) 16 15 Tittabawassee, below Dow dam 39 (13) 17 15 Tittabawassee, below Dow dam 83 (44) 35 15 Tittabawassee, below Dow dam n.d. (54) 18 16 Tittabawassee, at Consumer's Power Plant 301 (13) 19 16 Tittabawassee, at Consumer's Power Plant 129 (21) 20 17 Tittabawassee, at Freeland 66 (13) 21 18 Tittabawassee, below Saginaw Twp. Sewage Treatment Plant 64 (56) 22 18 Tittabawassee, below Saginaw Twp. Sewage Treatment Plant 125 (52) 23 18 Tittabawassee, below Saginaw Twp. Sewage Treatment Plant 135 (10) 34 19 Tittabawassee, at Wicke's Park n.d. (35) 24 19a Saginaw River, below Saginaw Sewage Treatment Plant 319 (190) 25 20 Saginaw Bay, near mouth of Saginaw R. 288 (105) 26 21 Saginaw Bay, near mouth of Saginaw R. 43 (26) 27 21 Saginaw Bay, near mouth of Saginaw R. 172 (92) 28 21 Saginaw Bay, near mouth of Saginaw R. 28 (8) 33 21 Saginaw Bay, near mouth of Saginaw R. n.d. (50) 37 22 Cass River, at Frankenmuth n.d. (40) 38 22 Cass River, at Frankenmuth n.d. (9) 39 23 Flint River, below Flint n.d. (100) 40 24 Flint River, Holloway Reservoir n.d. (24) 8 25 St. Clair River, at Decker's Landing 586 (81) Sample Number 124 Table 17 cont'd Sample Number Map I.D. Location 2,3,7,8-TCDD ng/kg 41 26 Clinton River at Avon Rd. n.d. (45) 42 27 Belle River at Gratiot Rd. n.d. (76) 62 28 Detroit River at Belle Isle n.d. (44) 43 29 Huron River at Superior Rd. n.d. (90) 44 30 Huron River at Plat Rock 246 (100) 7 30 Huron River at Plat Rock n.d. (50) 45 31 Raisin River at Monroe n.d. (52) 46 31 Raisin River at Monroe n.d. (24) 9 - Lake Erie at Port Clinton OH 75 (42) 59 - Lake Erie at Port Clinton OH n.d. (30) 60 - Lake Erie at Port Clinton OH n.d. (18) 61 - Lake Erie at Port Clinton OH n.d. (11) Number in parenthesis ( ) indicates the limit of detection of 2/3,7/8-TCDD for that specific sample based on the mass of internal standard present in the injection aliquot. The limit, of detection is reported as ng/kg 2,3,7,8-TCDD. LAKE MICHIGAN 125 Figure 8 . Map of Michigan Identifying Sample Sites 126 V. DISCUSSION A. Evaluation of the Analytical Technique and Results At the time of initiation of the research, the analytical technique employed was judged by a panel of independent consul­ tants as reliable for the determination of dioxin residues in biological media at the part per trillion level and representing the state of the art in trace analysis (185). The procedure evaluated was the one used in this work (183), with the exception that a packed column was used for gas chromatography. The packed column did not have the capability to resolve 2,3,7,8TCDD from a number of the other tetrachlorinated isomers of TCDD. The final recommendations of the panel stressed that specificity towards 2,3,7,8-TCDD needed to be achieved. Attainment of the chromatographic specificity for 2,3,7,8-TCDD was a major goal of this work. As evidenced by the results from the previous section, that goal was realized, providing satisfactory confirm­ ation of the residues as 2,3,7,8-TCDD. All of the samples with adequate recoveries were quantified on the basis of the total area of a completely resolved chroma­ tographic peak with a response at or above 2.5 times the noise level. No samples were quantified as part of a shoulder or multiple peak. A sample judged suitable for quantification re­ quired the presence of all four characteristic ions, eluting at the proper gas chromatographic retention time and with masses 319.9 and 321.9 at the correct isotope ratio (-25%). Only one of the procedural blanks was found to be contami- 127 nated by 2,3,7,8-TCDD. The contamination was traced to incomplete rinsing of the HPLC sample loop following calibration by 2,3,7,8TCDD. A record of the order in which samples were purified by HPLC was maintained. Additional instances of this type of con­ tamination were ruled out by closely inspecting the final GC/MS chromatograms of other samples which were processed following HPLC calibration. It turned out that most of these samples were not used due to low recoveries of internal standard. The absence of background contamination in the rest of the procedural blanks indicates with some certainty that the residues detected in the fish samples do not represent "false positive" values. A major problem encountered during analysis which seriously effected the interpretation of data was caused by poor sample recoveries. The validity of samples with quantifiable residues of native 2,3,7,8-TCDD was not conpromised by the poor recoveries. If the recovery of tracer is assumed to accurately reflect losses of native TCDD, any quantifiable recovery of the tracer can be used to calculate a corrected TCDD concentration. However problems are encountered when the internal standard is quanti­ fiable, but sample recovery is too low to allow detection of native 2,3,7,8-TCDD at a reasonable level of sensitivity. In such a case, the sample is determined to be negative for residues of 2,3,7,8-TCDD at a given limit of detection. The conclusion that the sample is negative for 2,3,7,8-TCDD at given limit of detection can be misinterpreted as indicating that the sample is entirely free of 2,3,7,8-TCDD. A sample with 10 ng/kg TCDD cannot be adequately compared to one which is negative at a 128 minimum detectable level of 100 ng/kg. The theoretical limit of detection of 1 ng/kg 2,3,7,8TCDD was not attained in any of the samples analyzed. Although the GC/MS sensitivity (20 pg) would have facilitated detection at this level, it requires that the sample be processed at 100% recovery. The average sample recovery achieved in this work was 4.5%, based on the amount of tracer present in the injection aliquot. A recovery of 100% requires extremely efficient cleanup and transfer steps. It is virtually impossible to extract and con­ centrate the entire residue into 2 ul of solvent for final GC/MS quantification. There is some confusion in the literature with respect to the reporting of sample recoveries. Many investiga­ tors report recoveries based on the final volume of extract, a portion of which is injected for GC/MS analysis. This can be done when you have a substantial amount of residue and utilize a larger mass of internal standard (ca. 20 ng ) . This scheme could not be utilized in this study because of low total recover­ ies and, due to limited amounts of tracer available, the incorporation of only 5 ng of 13 C-2,3,7,8-TCDD. Many of the problems encountered during this study were caused by attempting to process the samples before GC/MS confirm­ ation of the procedure could be performed. Analysis was carried out in stages, with the entire set of samples being processed together. Ideally, a batch of no more than 20 samples should be processed through the entire procedure, including GC/MS quanti­ fication, in as short a period of time as possible. In this way, problems caused by contamination, interferences or recoveries 129 can be immediately detected and corrected. The samples could not be processed in this manner for a number of reasons. Pro­ bably the major factor preventing the analysis in an integrated fashion was the limited availability of the necessary equipment and standards. The HPLC was utilized for a number of different projects and had to be modified for HPLC cleanup. Additionally, before quantification could £e -performed, the specificity of the procedure towards 2,3,7,8-TCDD had to be verified. This required the synthesis, identification and maximization of chromatographic resolution of the TCDD products by GC/MS. Problems encountered during the computer interfacing of our DuPont 321 GC/MS system and an unusually large number of major breakdowns rendered the GC/MS unavailable for use during most of the period of research. A decision was made to enter into a cooperative agreement with the M.S.U. Mass Spectrometry Facility for the limited use of a GC/MS. Use of a Hewlett- Packard 5985 GC/MS for a period of 4 weeks was arranged. During this time, the pyrolysis products were identified and resolved, mass spectrernetrie sensitivity was maximized and all of the sample extracts were quantified. Over 300 samples of fish were processed. 166 samples were determined to have suitable cleanup, as evidenced by their HPLC chromatograms. These 166 sample extracts were analyzed by GC/MS. Only 90 samples displayed recoveries suitable for quantification. Losses were most likely introduced during sample reconstitution and transfer and the result of adsorption, volatization and photodegradation during the long period of sample storage. some cases the extracts had been stored for over two years. In 130 Major adsorptive losses have been reported to occur if the samples are not analyzed immediately (185). Although the samples were processed according to established protocols and the utmost effort was expended to prevent the appearance of false positive values, there is still a slight possibility that the sample extracts may have contained materials with characteristic ions which could not be resolved from those of TCDD by low resolution mass spectrometry. The major inter­ fering ions arise from the presence and coelution of a component of Arochlor 1254 which has a characteristic ion of m/e 321.8679, by DDE ion at m/e 321.929 or by tetrachlorinated benzy phenyl ethers which display nominal mass fragments at 320, 322 and 324. However, the erroneous quantification of these interfering ions should have been eliminated by the column and HPLC cleanup steps which are reported to efficiently remove them at a 2 x 10^ fold excess (186). On this basis, a fish sample with a 20 mg/kg concentration of DDE would be sufficiently cleaned up to allow proper quantification of a sample with 10 ng/kg native 2,3,7,8TCDD without interference. Purification of the extracts by HPLC functioned to clean up the samples even further and separated the TCDDs from the other chlorinated dioxins. Additionally, the characteristic ions of TCDD are 500 times more intense than the interfering ions of DDE or PCB (187) and about 150 times as intense as those of the chlorinated benzyl phenyl ethers (186). resolution Although the capillary chromatographic of 2,3,7,8-TCDD from these interfering compounds has not been demonstrated in this work, it should provide an addition­ al safeguard against the introduction of false positive values 131 by some of these materials. Ultimate verification of the resi­ dues could be performed by use of a high resolution instrument. Confirmation of a portion of samples by high resolution GC/MS would be a valuable addition to a quality assurance program, but in most cases should not be a requirement of all routine analyses. Based on the author's experiences from this work, the following recommendations should be considered by persons attempt­ ing to perform dioxin analyses: 1) Analyses should be performed by a team of conscientious, patient technicians who are know­ ledgeable in all aspects of the procedure. 2) The team should be divided into four groups responsible for optimization of one facet of the procedure: a) extraction and cleanup b) HPLC purification c) GC/MS analysis d) reagent preparation and guality control. 3) Sufficient quantities of all 22 isomers of TCDD as well as the ^C and ^^C analogs of 2,3,7,8-TCDD must be available. 4) Samples should be processed with at least 10% procedural blanks and 5% duplication. 5) Samples should be processed and analyzed within the shortest time possible. 6) If possible, provisions should be made to verify at least 5% of the samples by high resolution GC/MS. 132 B. Geographic Distribution and Possible Sources of Residues The 2,3,7,8-TCDD residue levels listed in Table 17 agree with previous reports of TCDD in fish from the Tittabawassee and Saginaw rivers and Saginaw Bay (Table 10). Quantifiable levels of 2,3,7,8-TCDD were detected in almost all of the fish sampled downstream to the Dow-Midland facility. Samples of carp and sucker collected from the Tittabawassee and Chippewa rivers, eight km upstream to the Dow facility and samples collected from the Pine river, 40 km upstream to Midland also contained residues as high as those detected in samples originating below Midland. Presence of 2,3,7,8-TCDD in the upstream samples rules out aqueous effluent from Dow as the sole source of TCDD detected in the Saginaw watershed. Fish samples originating in entirely different watersheds, over 160 km from Midland were also determined to contain quanti­ fiable residues of 2,3,7,8-TCDD. The most significant aspect of the results of this survey is that 2,3,7,8-TCDD was detected in samples of fish collected from areas which did not contain trichlorophenol manufacturing sites. The detection of TCDD at these sites suggests that other processes may be contributing to the presence of the 2,3,7,8-TCDD residues. This study was initially designed to provide a data set of 2,3,7,8-TCDD determinations in fish from every major river in Michigan. However, a large number of samples were lost due to low recoveries. Samples with non-detectable TCDD residues at high limits of detection are also not suitable for direct comparison. The relatively small sample size remaining limits statistical 133 treatment of the data to a Z-test for proportions (188). This test is carried out on the basis of whether a given member of a population contains a quantifiable residue and does not address the magnitude of that residue. The data set can be divided into the following two populations: A) Samples collected within an 8 km radius of the Dow-Mid­ land facility and extending downstream into the Saginaw River and Saginaw Bay. B) Samples collected from other portions of the state. The proportions of samples in each population with quantifiable residues of 2,3/7,8-TCDD are compared aqainst the null hypothesis: Hq : There is no difference between the two populations. We can designate a sample with a residue level of greater than 50 ng/kg as "positive", while a sample with no detectable residues at or below a limit of detection of 50 ng/kg is class­ ified as "negative". According to this criterion, any sample which contains a quantifiable residue below 50 ng/kg is treated as a "negative" observation, while samples with limits of detec­ tion above 50 ng/kg which do not have quantifiable residues of TCDD are not included in the test. There were 12 "positive" samples in a total of 19 qualifying samples comprising the "A" population (sample numbers 10, 13 through 28, 33, 34 from Table 17 and 18). The remaining 30 qualifying samples (sample numbers 1 through 9, 11,12,30,31,32, 37,38,40,41,44,47,48,50,52,53,57 through 62) comprise the "B" or control population. This population contained 11 "positive" residues. The value of Z calculated by comparison of these two popula- 134 tions is 1.80. This value allows rejection of the null hypothesis at a level of confidience of 0.05. We can conclude that there is a significant difference between the two populations, with a greater proportion of positive samples appearing in the population of samples collected from the Midland area extending into Saginaw Bay. The following calculations and formulas were used: Z = ( Pi 1 “ P o ) ” < P i " P o ) p* x q* x ( 1/n^ + l/n2 ) P * = X 1 * X2 q* = 1 - p* nl + n2 p^ = proportion of positive samples in "A” population = 12/19 p£ = proportion of positive samples in "B" population = 11/30 n^ = total number of samples in "A" population = 19 n2 = total number of samples in "B" population = 30 V Pi ' ?2 = 0 Calculated value of Z = 1.80 Z0.05 " L '64 Conclusion: >P2 ; Reject null hypothesis. Aside from this somewhat weak Z-test, the data does not hold any other statistical significance. Sample sizes at each of the sites are too small to predict the level or presence of 2,3,7,8-TCDD in additional samples of fish. Samples originating outside of the Midland area can only be evaluated on an individual basis. The detection of residue of 2,3,7,8-TCDD does not indicate that a particular river is "polluted" by dioxins. It does 135 indicate that there may be a source near the site or somewhere in the watershed which can result in the establishment of residues of 2,3,7,8-TCDD in fish. Although one can only speculate as to the sources of the dioxins in the Tittabawassee and Saginaw Rivers, circumstances would render the chlorophenol production facility the most ob­ vious contributor. There is no question as to the potential for formation of dioxins during synthesis of chlorophenols by the alkaline hydrolysis method; even as associated with a modern trichlorophenol reactor (107). Production of 2,4,5-trichlorophenol has been carried out in Midland for a number of decades. Process design, standards of quality control and accepted practices of waste disposal have changed dramatically during this period. The TCDD could have entered the environment around Midland with aqueous wastes or through aerial deposition as a result of inefficient incineration of reactor wastes as well as through the formation of new TCDD from chlorophenols present in the wastes being combusted. The disposal of chlorophenol wastes by landfilling is another potential source of the environmental residues of TCDD. A single container of still bottom from an early chlorophenol reactor which produced high levels of 2,3,7,8-TCDD has the potential to cause widespread contamination at the part per trillion level if disposed of improperly. al are available. No records of chlorophenol waste dispos­ Independent waste haulers have been known to dispose of chemical wastes by various means, including application to rural roadways for dust control. It is possible that some of the dioxin-containing wastes were landfilled at various sites 136 throughout the state and are contributing to the residues de­ tected at sites far removed from Midland. The formation of dioxins during combustion is also a valid potential source of environmental residues of 2,3,7,8-TCDD. Studies cited in the previous sections clearly indicate that significant amounts of dioxins are produced during combustion of municipal waste. An estimated 77 ug of 2,3,7,8-TCDD can be produced during the incineration of a ton of municipal waste (see page 50 for reasoning and citations). This value would be expected to vary, depending on the type of waste and conditions in the incinerator, but it is presently the best estimate avail­ able. On the basis of information derived from air emission permits issued by the Air Quality Division of the Michigan Depart­ ment of Natural Resources, a total of 400,000 tons of municipal waste, 800,000 tons of dry sewage sludge and 40,000 tons of liquid chemical waste is disposed of by incineration annually in Michigan. An additional 500,000 tons of waste is burned in smaller, privately-owned incinerators. The total of 1,740,000 tons of waste could result in the formation of 134 grams of 2,3,7,8-TCDD yearly. The amount of additional 2,3,7,8-TCDD entering the environ­ ment through the combustion of fossil fuels is difficult to estimate. There are no data available on the yields of dioxins produced during the burning of coal for heating and power gener­ ation. If levels similar to those encountered during waste incineration are produced, these processes could be providing a major input of TCDD into the environment. Results of laboratory- scale combustion experiments suggest that dioxins are formed 137 during the burning of coal, but no studies have demonstrated the presence of 2,3,7,8-TCDD in fly ash or airborne emissions from coal-fired power plants. One could conclude that dioxins are being produced during the burning of coal on the basis of reaction schemes presented in Figure 4. The 134 grams of 2,3,7,8-TCDD estimated as being produced in Michigan during waste incineration is equivalent to the amount of 2,3,7,8-TCDD present in 4.4 metric tons of 2,4,5-T with an average 2,3,7,8-TCDD level of 30 mg/kg, or 1340 metric tons of 2,4,5-T with an average concentration of 2,3,7,8-TCDD of 0.1 mg/kg. About 3400 metric tons of 2,4,5-T were applied in the U.S. in 1974. If a similar amount were applied prior to 1971, it could represent an annual nationwide 2,3,7,8-TCDD input ranging from 0.340 to 102 kg, assuming respective TCDD levels in 2,4,5-T ranging from 0.1 to 30 mg/kg. Use of 2,4,5-T has since been suspended, but synthesis of ortho chlorophenols continues, pri­ marily due to the extreme usefulness of these important chemical intermediates. If one assumes that dioxin generation during chlorophenol production is currently being minimized and carefully monitored, inputs from combustion can now be considered major contributors to residues of 2,3,7,8-TCDD. The entry of 2,3,7,8-TCDD to the environment from combustion processes can be minimized by efficient particulate collection and the proper landfilling of fly ash. The fly ash may contain levels of other toxic chlorinated dioxins and dibenzofurans which render its toxic potential over 10 times greater than based on the level of 2,3,7,8-TCDD alone (142)• Failure to con­ fine fly ash may result in its distribution in the environment, 138 resulting in the biological attenuation of the chlorinated dioxins and furans adsorbed onto it. The results of this work demonstrate the need for continued research into the formation of dioxins during combustion and its impact on the appearance of environmental residues of 2,3/7,8TCDD. A critical question to be addressed is whether coal-fired power plants produce dioxins at significant levels. Studies into the presence of TCDD in aquatic sediments and soils should also be conducted to establish background levels of TCDD around the state. The residue values are especially difficult to interpret because of the variety of potential dioxin sources in Michigan. A similar study should be conducted in an area which contains an industry that relies heavily on fossil fuel energy but does not support a chlorophenol facility. The significance of the residues detected is also tenuous because they are among the first observations conducted in the environment for a specific compound at the part per trillion (ng/kg) level. At this level of sensitivity/ many compounds previously undetected and thought to be absent from certain portions of the biosphere appear with apparent ubiquity. 139 C. Ecological and Public Health Significance of the Residues This study was primarily concerned with detecting the presence of 2, 3,7,8-TCDD in aquatic environments. Fish were chosen as a test animal because they are very good indicators of environmental contamination by recalcitrant, lipophilic or­ ganic compounds. They can accumulate many of these materials to levels orders of magnitude higher than generally present in the environment. Fish attenuate TCDD biologically and represent an organism which is a human food source. They can be easily sampled and contain enough tissue for multiple analyses. Carp and sucker were chosen as test species because they feed on organisms and materials primarily associated with mud and sediment. Sediment has a high adsorption affinity towards 2,3,7,8-TCDD and can represent a major reservoir for dioxins in the aquatic environment. Because carp and sucker cycle so much sediment and bottom-dwelling organisms, they are expected to con­ tain higher levels of TCDD than other species of fish. Available studies of the cycling of 2,3,7,8-TCDD in model 3 ecosystems report a fish concentration factor of about 10 , calculated relative to the concentration of TCDD present in the water column (168). Because we are assuming that carp and sucker derive most of the TCDD from sediment during feeding, calculation of a water concentration for TCDD based on the resi­ due levels detected in fish should result in a high approximation. The range of TCDD residue values measured in this work was 17 to 586 ng/kg. Corresponding estimates of water levels based on the model ecosystem data range from 0.017 to 0.588 ng / 1 . 140 Based on the results of static fish bioassays, concentrations of 2,3,7,8-TCDD in the water column within the estimated range can be toxic to fish. A water level of 1.0 ng/1 TCDD caused complete mortality of mosquito fish (39) . Exposure of young coho salmon for 96 hours at a water concentration of 5.6 ng/1 resulted in 55% mortality within 80 days postexposure, while water levels as low as 0.056 ng/kg (0.054 ng/g body weight) TCDD for 96 hours resulted in a 10% increase in mortality (189). Exposure of developing yolk sac fry and juvenile rainbow trout and pike to a static water concentration of 0.100 ng/kg 2,3,7,8TCDD resulted in decreased embryonic development, decreased growth and increases in skeletal malformations as well as death (36). These studies demonstrate that very low levels of 2,3,7,8TCDD in water are capable of inducing deleterious effects in exposed populations of aquatic organisms. Although estimates of the water concentration may be high, the residues detected in this study indicate that TCDD may be present in certain rivers in Michigan at concentrations near these action levels. Fish can also function as a source of TCDD to terrestrial populations which use them as a source of food. An experiment to assess the magnitude of TCDD taken up by mink fed a diet of carp originating in Saginaw Bay was conducted as part of this research. The mink were fed ground Saginaw Bay carp for at least 30 days as part of an experiment to assess the impact of PCBs present in the carp on mortality and reproductive success (190). Samples of the feed and livers from the exposed mink were provided, by the investigators conducting the bioassay. Table 19 lists the levels of 2,3,7,8-TCDD detected in he mink livers. All livers contained 141 detectable residues of 2,3,7,8-TCDD ranging from 39 to 87 ng/kg. A sample of pooled livers from the control population and samples of the feed were not suitable for comparison due to low recoveries, An average residue of 57 ng/kg in a mean liver weight of 12 g represents a total mass of 684 pg of 2,3,7,8-TCDD per liver. The extent of accumulation cannot be evaluated without knowledge of the average TCDD content of the feed. The 684 pg of 2,3,7,8- TCDD could have been provided to the mink by ingestion of a total of 34 g of carp with a mean 2,3,7,8-TCDD content of 20 ng/kg. Table 18 2,3,7,8-TCDD in Livers of Mink Fed Carp from Saginaw Bay . o «7 o / /, \ Limit of detection Sample________________ 2,3,7,8-TCDD (ng/kg)____________ (ng/kg)_______ A 39 12 B 87 8 C 46 23 An epidemiological assessment of the hazards associated with the ingestion of TCDD-contaminated fish has been conducted by researchers at the U.S. Food and Drug Administration (191). They used data from a 2-year rat feeding study (20) as the basis for determining human hazards associated with the ingestion of small quantities of 2,3,7,8-TCDD present in fish. The chronic rat feeding study reported no observable effects to rats at a life­ time exposure of 1 ng/kg body weight per day; hyperplasia of the liver and lung cells at a dose level of 10 ng/kg-day and an in­ crease in liver carcinomas at a dose level of 100 ng/kg-day. 142 The corresponding 2,3,7,8-TCDD doses to a 70 kg human subject would be 70, 700 and 7000 ng/day. A comprehensive fish consumption study gathered information from about 25,000 individuals in the Great Lakes area. A ranked distribution of those persons who consumed fish from the Great Lakes, including inland lakes and rivers, revealed that a person in the 99th percentile (those consuming the most fish) consumed 36.8 g of fish daily, or 257.6 g per week. An individual in the 90th percentile consumed an average of 15.7 g of fish per day (109.7 g/week). An individual in the 99th percentile consuming 36.8 g of fish daily with 2,3,7,8-TCDD consistently at the highest level detected in this study (586 ng/kg) would be ingesting TCDD at a dosage of 0.31 ng/kg body weight per day, or a dose of 21.6 ng/day. This dose is less than one-third of the 70 ng/day no­ effect level. If the TCDD level was equal to the average positive residue level detected in this survey (178 ng/kg) the individual would be ingesting 6.5 ng TCDD/day, which is less than 10% of the no-effect level. The no-effect level would not be exceeded unless the concentration of 2,3,7,8-TCDD in the fish consistently exceeded 1902 ng/kg, or 1.9 ppb. A fish consumer belonging to the 90th percentile would have to eat 15.7 g of fish daily with an average TCDD level of greater than 4.4 ug/kg (4.4 ppb) 2,3,7,8- TCDD in order to exceed the no-effect level. At the 99th percentile, the concentration required to induce hepatic injury would be 19 ug/kg. The TCDD present in the fish would have to exceed 190 ug/kg to reach the carcinogenic effect level. 143 These estimates are based on the individual consuming fish with the respective TCDD residues daily over his entire lifetime. The probability of every meal of fish exceeding the threshold concentration is very low. Few fish would be able to survive a body burden as high as 1.9 ug/kg 2,3,7,8-TCDD. An additional margin of safety would be realized following cooking of the fish. Table 20 lists the results of cooking experiments conducted on samples of fish with known concentrations of 2,3,7,8-TCDD. Reductions in TCDD content ranged from 26 to 67% following broiling. The reductions were calculated relative to the final weight of the cooked fish. If the loss of moisture (ca. 20%) is considered, the reductions in TCDD content relative to the raw fish are even greater. Table 19 TCDD Residues in Fish Before and After Broiling 2,3,7,8-TCDD in fish following broiling, ng/kg 2,3,7,8-TCDD in raw fish, ng/kg % reduction 136 (36) 45 (24) 67 83 (44) 61 (38) 26 64 (56) 36 (18) 44 287 (84) 141 (11) 51 ( ) = limit of detection of TCDD A more conservative estimate of the hazard associated with ingestion of 2,3,7,8-TCDD contaminated fish can be extrapolated from the minimum cumulative toxic dose of 0.1 ug/kg calculated 144 by Stevens (192). His calculations are based on observations made on human subjects suffering from Yusho disease, induced by 2,3,7,8-tetrachlorodibenzofuran. He reasons that since TCDF is 20 times less toxic than 2,3,7,8-TCDD in most species, the minimum toxic dose of 2 ug 2,3,7,8-TCDF/kg should be equivalent to 0.1 ug/kg 2,3,7,8-TCDD, or 7 ug of 2,3,7,8-TCDD in a 70 kg human. An individual in the 99th percentile, consuming 36.8 g of fish daily with the highest detected residue of 586 ng/kg would exceed the toxic threshold in about 324 days. If the level of TCDD in the fish was 178 ng/kg (average positive level observed), it would take 1077 days to reach the threshold, while consumption of fish with a TCDD residue of 20 ng/kg would require 26 years to exceed the threshold. These calculations neglect mechanisms of excretion and metabolism which would further prevent the effective dose from being reached. They also neglect the safety factor introduced during cooking. On the basis of these calculations, humans should not be significantly effected by eating reasonable amounts of fish containing 2,3,7,8-TCDD. However, the margin of safety realized by humans as a result of omnivorous eating habits, cooking practices and a relatively high body weight would not apply to wildlife species that consume fish as a major component of their diet. Assuming equal sensitivity to TCDD, a 5 kg raccoon con­ suming 100 g of fish per day with an average 2,3,7,8-TCDD level of 500 ng/kg can exceed the 0.1 ug/kg threshold in 10 days. A 750 g mink ingesting 50 g of fish daily with a residue level of 500 ug/kg would take in 25 ng TCDD per day and reach the toxic 145- dose in 3 days. Ingestion of 50 g of fish daily with an average TCDD level of 150 ng/kg would result in a dosage of 10 ng/kg-day and exceed the toxic dose in 8 days. Fish-eating birds such as diving ducks and gulls would be exposed to similar hazards from the consumption of dioxin-contaminated fish. The potential stresses imposed upon populations of wild­ life as a result of consumption of dioxin-contaminated fish is considerably greater than those imposed upon the human popula­ tion. This hazard would be magnified by the presence of other toxic compounds in the fish, such as PCBs, chlorinated dibenzofurans and other chlorinated dioxin congeners. The result of the stresses can be a reduction in the reproductive capacity of the exposed individuals, increased mortality, impaired predator avoidance behavior and a number of other factors which could ultimately cause a decline in the size of the population. The calculated assessments of the risk posed to wildlife as a result of environmental contamination are based on toxicity data from different species and on a few observations of TCDD residues present in fish. Further investigations into the concen­ trations of chlorinated dioxins and dibenzofurans present in fish and the inclusion of fish-eating wildlife species in residue studies will result in a better understanding of the potential hazards of environmental residues of these materials. Investi­ gation of the acute and chronic toxicity of dioxins to wildlife should be conducted to assess the sensitivity of those species to the toxic effects of 2,3,7,8-TCDD. 146 VI. CONCLUSIONS The analytical method utilized in this research can be successfully used to quantify residues of 2,3,7,8-TCDD in fish at the part per trillion (ng/kg) level with absolute chromato­ graphic specificity towards 2,3/7,8-TCDD. The technique is difficult to carry out and requires the availability of extensive dedicated instrumentation. A strict quality control program should be maintained to detect and prevent contamination which may result in high background values or the appearance of interT O fering compounds. Standards of T O C-, T C- and A C - 2 ,3,7,8-TCDD as well as the 21 other isomers of TCDD are crucial to method validation and final sample quantification. Poor sample recoveries result in decreasing the sensitivity of the procedure. Recovery can be maximized through careful sample evaporations and transfers. Processing and quantifying the samples in as short a period of time as possible minimizes volatization and adsorption of TCDD onto glassware during storage. The results indicate that residues of 2,3,7,8-TCDD ranging from 17 to 586 ng/kg can be detected in fish from many rivers in Michigan which flow through industrialized urban areas. A great­ er proportion of samples containing quantifiable levels of 2,3,7,8TCDD originated in the Tittabawassee and Saginaw Rivers, down­ stream to Midland. These residues are presumably the result of the. large amount of chlorophenol manufacturing which has taken place in Midland over many years. However, the formation of dioxins during combustion of municipal wastes and fossil fuels cannot be ruled out as contributing to the existence of the 147 residues. Additional toxic dioxin and dibenzofuran congeners can be produced during combustion, which may significantly increase the potential hazard of combustion as a dioxin source. Studies into the formation of chlorinated dioxins and dibenzofurans in coal-fired power plants must be intensified to properly evaluate the magnitude of combustion as a source of environmental residues of TCDD. The concentration at which 2,3,7,8-TCDD was detected in fish suggest that 2,3,7,8-TCDD could be present in the aquatic envi-: ronment at levels potentially toxic to aquatic organisms. The residues detected in fish should not pose a human health risk to individuals consuming fish in moderate amounts. An added mar­ gin of safety is gained during cooking, which reduces TCDD content. However, wildlife which normally consumes large amounts of fish may be exposed to potentially harmful doses of 2,3,7,8-TCDD and other toxic materials. LITERATURE CITED 148 Literature Cited 1. Poland, A. and Knutson, J. 19 82. 2,3,7,8-tetrachlorodibenzo-p-dioxin and related halogentaed aromatic hudrocarbons? Examination of the mechanism of toxicity. Ann. Rev. Pharmacol. Toxicol. 22:517-24 2. Schwetz, B., Norris, J., Sparschu, G., Rowe, V., Gehring, P., Emerson, J.L. and Gerbig, C. 1973. Toxicology of chlorinated dibenzo-p-dioxins. Environ. H l t h . Perspect. 5:87-99 3. Vos, J., Moore, J. Zinkl, J. 1973. Effects of 2,3,7,8terachlorodibenzo-p-dioxin on the immune system of laboratory animals. Environ. Health Perspect. 5:149-62 4. Harris, M . , Moore, J., Vos, J., Gupta, B. 1973. General biological effects of TCDD in laboratory animals. Environ. Health Perspect. £: 101-9 5. McConnel, E., Moore, J., Haseman, J., Harris, M. 1978 The comparative toxicity of chlorinated dibenzo-p-dioxins in mice and guinea pigs. Toxic. Appl. Pharmacol. 44:335 6. McConnel, E., Moore, J., Dalgard, D. 1978. Toxicity of 2,3,7,8-tetrachlorodibenao-p-dioxin in Rhesus Monkeys (Macaca mulatta) following a single oral dose. Toxicol. A p p l . Pharmacol. 43(1):175-87 7. Beatty, P., Vaughn, W., Neal, R. 1978. Effect of alteration of rat hepatic mixed function oxidase (MFO) activity on the toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Toxicol. A p p l . Pharmacol. £5(2):513-19 8. Allen, J., Van Miller, J., Norback, D. 1975..Tissue distri­ bution, excretion and biological effects of C-tetrachlorodibenzo-p-dioxin in rats. Food Cosmet. Toxicol. 13:501-5 9. Neal, R . , Olson, J., Gasiewicz, T., Geiger, L. 1982. The toxicokinetics of 2,3,7,8-tetrachlorodibenzo-p-dioxin in mammalian systems. Drug Metab. R e v . 13(3):355-85 10. Jones, G., and Greig, J. 1975. Pathological changes in the liver of mice given 2,3,7,8-tetrachlorodibenzo-p-dioxin. Experientia 31:1315-17 11. Vos, J., Moore, J., Zinkl, J. 1974. Toxicity of 2,3,7,8terachlorodibenzo-p-dioxin (TCDD) in C57B1/6 mice. Toxicol. Appl. Pharmacol. 29:229 149 12. Henck, J., New, M., Kociba, R., Rao, K. 1981. 2,3,7,8tetrachlorodibenzo-p-dioxin: acute oral toxicity in hamsters. Toxicol. A p p l . Pharmacol. 59:405-7 13. Olson, J.R., Holscher, M.A., Neal, R.A. 1980. Toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the Golden Syrian Hamster. Toxicol. Ap p l . Pharmacol. 55:67-78 14. Cunningham, H.M., and Williams, D.T. 1972. Effect of tetrachlorodibenzo-p-dioxin on growth rate and the synthesis of lipids and proteins in rats. B u l l . Environ. Contarn. Toxicol. 7:45-51 15. Fowler, B.A., Lucier, G.A., Brown, H.W., Bullock, B.C. 1975. Ultrastructural changes in rat liver cells following a single oral dose of TCDD. Environ. Health Perspec. 1:141-48 16. Greig, J.B., Jones, G., Butler, W.H., Barnes, J.M. 1973. Toxic effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Food Cosmet. Toxicol. 11:585-95 17. Gupta, B.N., Vos, J.A., Moore, J.G., Zinkl, J.G., Bullock, B. 1975. Pathologic effects of 2,3,7,8-tetrachlorodibenzo-pdioxin in laboratory animals. Environ. Health Perspect. 5:125-40 18. Jones, G. 19 75. A'histochemical study of the liver lesion induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin (dioxin) in rats. J. Pathol. 116;101-5 19. Kociba, R.J., Keeler, P.A., Park, C.N., Gehring, P.J. 1976. 2.3.7.8-tetrachlorodibenzo-p-dioxin (TCDD). Results of a 13-week oral toxicity study in rats. Toxicol. A p p l . Pharmacol. 35:553-74 20. Kociba, R.J., Keyes, D.G., Carreon, R.M., Wade, C.E., Dittenber, D.A., Kainins, R.F., Frauson, L.E., Park, C.N., Hummel, R.A., Humiston, C.G. 1978. Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8-tetra­ chlorodibenzo-p-dioxin in rats. Toxicol. A p p l . Pharmacol. 15(1):298 21. Murray,F.J., Smith, F.A., Nitschke, K.D., Humiston, C.G., Kociba, R.J., Schwetz, B.A. 1979. Three-generation repro­ duction study of rats given 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD) in the diet. Toxicol. A p p l . Pharmacol. 50(2): 241-52 22. Van Miller, J.P., Lalich, J.J., Allen, J.R. 1977. Increased incidence of neoplasms in rats exposed to low levels of 2.3.7.8-tetrachlorodibenzo-p-dioxin. Chemosphere 1:537-44 150 23. Faith, R.E., Moore, J.A. 1977. Impairment of thymus-depend­ ant immune functions by exposure of developing immune system to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). J. Toxicol. Environ. Health ,3(3) :451-64 24. Thigpen, J.E., Faith, R.E., McConnell, E.E., Moore, J.A. 1975. Increased susceptibility to bacterial infection as a sequela of exposure to 2,3,7,8-tetrachlorodibenzo-pdioxin. Infect. Immun. 12:1319-24 25. Weisberg, J.B. and Zinkl, J.G. 1975. Effects of 2,3,7,8tetrachlorodibenzo-p-dioxin upon hamsters and hematologic function in the rat. Environ. Health Perspect. J5:119-23 26. Zinkl, J.G., Vos, J.G., Moore, J.A., Gupta, B.N. 1975 Hematologic and clinical chemistry effects of 2,3,7,8tetrachlorodibenzo-p-dioxin in laboratory animals. Environ. Health Perspect. 5:111-23 27. Hinsdill, R.D., Couch, D.L., Speirs, R.S. 1980. Immuno­ suppression in mice induced by dioxin (TCDD) in feed. J. Environ. Path. Toxicol. £(2-3):401-25 28. Becker, D. 1973. The effect of folate overdose and of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) on kidney and liver, respectively, of rat and mouse embryos, (abstract) Teratology j3:215 29. Norback, D.H. and Allen, J.R. 1975. Biological responses of the nonhuman primate, chicken and rat to chlorinated dibenzo-p-dioxin ingestion. Environ. Health Perspect. 5:233-40 30. Seefeld, M.D., Yang, K.H., Peterson, R . E . 1979 Effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin on liver function in rats, rabbits and guinea pias. Toxicol. Appl. Pharmacol. 48(1):A153 31. Swift, L.L., Gasiewicz, T.A., Dunn, D.G., Soule, P.D., Neal, R.A. 19 81. Characterization of the hyperlipidemia in guinea pigs induced by 2,3,7,8 TCDD. Toxicol. A p p l . Pharmacol. 59:489-99 32. Allen, J.R., Barsotti, D.A., Lalich, J.J., VsnMiller, J.P., Abrahamson, L.J. 1977. Morphological changes in monkeys consuming a diet containing low levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Food Cosmet. Toxicol. 15(5) :401-10 33. Schantz, S.L., Barsotti, D.A., Allen,J.R. 1979. Toxicological effects produced in nonhuman primates chronically exposed to fifty parts per trillion 2,3,7,8-tetrachloro­ dibenzo-p-dioxin (TCDD). Toxicol. A p p l . Pharmacol. 48(1):A180 151 34. Milnes, M.H. 1971. Formation of 2, 3,7,8-tetrachlorodibenzop-dioxin by thermal decomposition of sodium 2,4,5-trichlorophenate. Nature (London) 232:395-6 35. Miller, R.A., Norris, L.A., Hawks, C.L. 1975. Toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in aquatic organisms. Environ. Health Perspect. 5^:177-86 36. Helder, T. 1982. Effects of 2,3,7,8 TCDD on early life stages of two fresh-water fish species. Pergamon S e r . Environ. S c i . 5^:455-62 37. Miller,R., Norris, L., Loper, B. 1979. The response of Coho Salmon and Guppies to 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD) in water. Trans. Am. Fish S o c . 108:401-7 38. "'"Norris,L. and Miller, R. 1974. The toxicity of 2,3,7,8-tetra­ chlorodibenzo-p-dioxin (TCDD) in guppies (Poecilia reticulatus Peters) Bull . Environ. Contam. Toxicol. 12(1):76-80 39. Yockim, R., Isensee, A., Jones, G. 1978. Distribution and toxicity of TCDD and 2,4,5-T in an aquatic model ecosystern.'""Chemosphere 3 ^ : 2 1 5 - 2 0 40. Flick, D.F., Firestone, D., Ress, J., Allen, J.R. 1973. Chick edema disease 10. Toxicity of chick edema factors in the chick, chick embryo and monkey. Poult. S ci. 52:1637-41 41. Fries, G.F. and Marrow, G.S. 1975. Retention and excretion of 2,3,7,8-tetrachlorodibenzo-p-dioxin by rats. J. A g r i . Food Chem. 23(2):265-69 42. VanMiller, J.P., Marlar, R.J., Allen, J.R. 1976. Tissue distribution and excretion of tritiated tetrachlorodibenzop-dioxin in non-human primates and rats. Food Cosmet. Toxicol. 14:31-4 43. Piper, W.N., Rose, J.Q., Gehring, P.J. 1975. Excretion and tissue distribution of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Environ. Health Perspect. 5^241-44 44. Williams, D*T., Cunningham, H.M., Blanchfield, B.J. 1972 Distribution and excretion studies of octachlorodibenzo-pdioxin in the rat. Bull. Environ. Contam. Toxicol. 2 :57-62 45. Moore, J.A., Harris, M.W., Albro, P.N. 1976. Tissue distrbution of 14C tetrachlorodibenzo-p-dioxin in pregnant and neonatal rats, (abstract). Toxicol. App l . Pharmacol.37:46 46. Piper, W.N., Rose, J.Q., Gehring, P.J. 1973. Excretion and tissue distribution of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Advan. Chem. Ser. 120:85-91 152 47. Tulp, M.T.,and Hutzinger, 0. 1978. Rat metabolism of poly­ chlorinated dibenzo-p-dioxins. Chemosphere 2 ;761-68 48. Ramsay, J.C., Hefner, J.G., Karbowski, R.J., Braun, W.H., Gehring, P.J. 1979. The in vivo biotransformation of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) in the rat. Toxicol. Ap p l . Pharmacol. 48^(1) :A162 49. Rose, J.Q., Ramsey, J.C., Mentzler, T.H., Hummel, R.A., Gehring, P.J. 1976. The fate of 2,3,7,8-tetrachlorodibenzop-dioxin following single and’repeated oral doses to the rat. Toxicol. A p p l . Pharmacol. 36:209-26 50. Vinopal, J.H. and Casida, J.E. 1973. Metabolic stability of 2,3,7,8-tetrachlorodibenzo-p-dioxin in mammalian liver microsomal systems and in living mice. A r c h . Environ. Contam. Toxicol. 1(2):122-32 51. Nolan, R.J., Smith, F.A., Hefner, J.G. 1979. Elimination and tissue distribution of 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD) in female guinea pigs following a single oral dose. Toxicol. A p p l . Pharmacol. 48(1):A162 52. Gasiewicz, T.A. and Neal, R.A. 1979. 2,3,7,8-tetrachlorodibenzo-p-dioxin tissue distribution, excretion and effects on clinical chemical parameters in guinea pigs. Toxicol. A p p l . Pharmacol. 51:329-39 53. Olson, J.R., Gasiewicz, T.A., Neal, R.A. 1980. Tissue distribution, excretion and metabolism of 2,3,7,8-tetra­ chlorodibenzo-p-dioxin in the Golden Syrian Hamster. Toxicol. A p p l . Pharmacol. 56(1):78-85 54. Neal, R.A., Olson, J.R., Gasiewicz, T.A., Geiger, L.E. 1982 The toxicokinetics of 2,3,7,8-tetrachlorodibenzo-p-dioxin in mammalian systems. Drug Metab. Review 13(3):355-85 55. Olson, J.R. 1981. Ref. in Toxicologist 1:69 56. Poiger, H. and Schlatter, C. 1980. Influence of solvents and adsorbents on dermal and intestinal absorption of TCDD. Food Cosmet. Toxicol. 18(5): 477-81 57. Sawahata, T. Olson, J.R., Neal, R.A. 1982. Identification of metabolites of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) formed on incubation with isolated rat hepatocytes. Biochem. Biophys. R e s . C o m m . 1 0 5 (1):344-46 58. Poiger, H., Buser, H.R., Weber, H., Zweifel, U .,Schlatter, C. 1982. Structure elucidation of mammalian TCDD metabolites. Experientia 38:484-86 59. Poiger, H., Weber, H. Schlatter, C. 19 82. Special aspects of metabolism and kinetics of TCDD in dogs and rats. Assesment of toxicity of TCDD metabolites in guinea pigs. Pergamon Ser. Environ. S c i . 153 60. Poland, A. and Glover, E. 1974. Genetic expression of aryl hydrocarbon hydroxylase activity. J. B i o l . C h e m . 249(17):5599-606 61. Poland, A., Glover, E., Kende, A. 1976. Stereospecific high affinity binding of 2,3,7,8-tetrachlorodibenzo-pdioxin by hepatic cytosol. Evidence that the binding species is a receptor, for induction of aryl hydrocarbon hydroxylase. J. B i o l . C h e m . 251:4936-46 62 Goldstein, J.A. 1980. Structure-activity relationships for the biochemical effects and the relationships to toxi­ city. In: Halogenated Biphenyls, Terphenyls, Naphthalenes, Dibenzodioxins and Related Compounds. Kimbrough, R.D., ed. pp 151-90. New York: Elsevier/North Holland. 406 pp. 63. Bradlaw, J., Garthoff, L., Hurley, N. 1980. Comparative induction of aryl hydrocarbon hydroxylase activity in vitro by analogues of dibenzo-p-dioxin. Food Cosmet. Toxicol. 18(6):627-35 64. Me Kinney, J.N., McConnel, E. 1982. Structural specificty and the dioxin receptor. Pergammon S e r . Environ. S c i . 5:367-81 65. Poland, A., Greenlee, W . , Kende, A. 1979. Studies on the mechanism of action of the chlorinated dibenzo-p-dioxins and related compounds. A n n . N.Y. A c a d . S c i . 320:214-30 66. Cheney, V., 1982. Structural factors'.: affecting aryl hydro­ carbon hydroxylase induction by dibenzo-p-dioxins and diben-. zofurans. Int. J. Quantum C h e m . 21:445-62 67. Collins, T.F. and Williams, C.H. 1977. Teratogenic studies with 2,4,5-T and 2,4-D in the hamster. Bul l . Environ. Contam. Toxicol. 6X6) :559-67 68. Jensen, D.J., Hummel, R.A., Mahle, N.H., Kocher, C.W., Higgins, S.H. 1981. A residue study of beef cattle consum­ ing 2,3,7,8-tetrachlorodibenzo-p-dioxin. J. A g r i . Food Che m . 29(2): 265-68 69. Courtney, K.D. and Moore, J.A. 19 71. Teratology studies with 2,4,5-trichlorophenoxyacetic acid and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. A p p l . Pharmacol. 20:396-403 70. Moore, J.A., Gupta, B.N., Zinkl, J.G., Vos, G. 1973. Post­ natal effects of maternal exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Environ. Health Perspect. 5:81-5 71. Moore, J.A. 1976. A Reference in Toxicol. A p p l . Pharmacol♦ 37:131 154 72. Neubert, D. and Dillman, I. 1972. Embryotoxic effects in mice treated with 2,4,5-trichlorophenoxyacetic acid and 2.3.7.8-tetrachlorodibenzo-p-dioxin. Naunyn Schmiedbergs A r c h . E x p . Pathol. Pharmacol. 272:243-64 73. Sparschu, G.L., Dunn, F.L., Rowe, V.K. 1971. Study of the teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Food Cosmet. Toxicol. j): 405-12 74. Vos, J.G. and Moore, J.A. 1974. Suppresion of cellular immunity in rats and mice by maternal treatment with 2.3.7.8-tetrachlorodibenzo-p-dioxin. Int. A r c h . Allergy A p p l . Imunol. 47:777-94 75. Dougherty, W.H., Coulston, F., Goldberg, L. 1973. Non­ teratogenicity of 2,4,5-trichlorophenoxyacetic acid in monkeys (Macaca mulatta). (Abstract) Teratology 8:215 76. Smith, F.A., Schwetz, B.A., Nitschke, K.D. 1976. Terato­ genicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in CF-1 mice. Toxicol. A p p l .Pharmacol. 38(3):517-23 77. Courtney, K.D. 1976. Mouse teratology studies with chlorodibenzo-p-dioxins. B u l l . Environ. Contam. Toxicol. 16(6):674-81 78. Khera, K.S. and Ruddick, J.A. 1973. Polychlorodibenzo-pdioxins: Perinatal effects and the dominant lethal test in Wistar rats. A d v . C h e m . S e r . 1 2 0 (8):70-84 79. Neubert, D., Zens, P., Rothenwalker, A., Merker, H.J. 1973. A survey of the embryotoxic effects of TCDD in mammalian species. Environ. Health Perspect. 5j67-79 80. Zingeser, M. 1979. Anomalous development of the soft palate in rhesus macaques (Macaca mullata) prenatally esposed to 2.3.7.8-tetrachlorodibenzo-p-dioxin. Teratology 19:54A 81. Giavini, E., Prati, M., Vismara, C. 1982. Rabbit teratology study with 2,3,7,8-tetrachlorodibenzo-p-dioxin. Environ. Research. 2J7 (1) :74-8 82. Remotti,G., Devirgiliis, F., Bianco, V., Candiani, G.B. 1981. The morphology of early trophoblast after dioxin poisoning in the Seveso area. Placenta 2_(l):53-62 83. Poland, A. and Gloxer, E. 1980. 2,3,7,8-tetrachlorodibenzop-dioxin: Segregation of toxicity with the Ah locus. Molec. Pharmacol. 17:86-94 84. Abate, L., Basso, P., Belloni, A., Bisanti, L., Borgna, C., Bruzzi, P., Dorigaotti, G., Falliva, L., Fanuzzi, A., Formigaro, M., Maggiore, G., Marni, E., Meazza, L., Merlo, F., Puntoni, R . , Rosa, A., Stagnaro, E., Vercelli, M. 1982. Mortality and birth defects from 1976 to 1979 in the popu­ lation livina in the TCDD oolluted area of Seveso. Perqamon --Series Environ. Sci. 5:571-87 155 85. Beatty, P.W., Lemback, K.J., Holscher, M.A., Neal, R.A. 1975. Effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on mammalian cells in culture. Toxicol. A p p l . Pharmacol. 31:309-12 86 . Green, S., and Moreland, F.S. 1975. Cytogenetic evaluation of several dioxins in the rat. Toxicol A p p l . Pharmacol. 33:161 87. Greig, J.B., Taylor, D.M., Jones, J.D. 1974. Effects of 2.3.7.8-tetrachlorodibenzo-p-dioxin on stimulated DNA synthesis in the liver and kidney of the rat. C h e m . Biol. Interact. J3:31-9 88 . Hussian, S., Ehrenberg, L., Lofroth, G., Gejvall, T. 1972. Mutagenic effects of TCDD on bacterial systems. Ambio .1:32-3 89. Jackson, W.T. 1972. Regulation of mitosis, 3. Cytological effects of 2,4,5-trichlorophenoxyacetic acid and dioxin contaminants in 2,4,5-T formulations. J. Cell Sc i . 10:15-25 90. Seiler, J.P. 1973. A survey of the mutagenicity of various pesticides. Experientia 29:622-23 91. Green, S., Moreland, F., Sheu, D. 1977. Cytogenetic effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin of rat bone marrow cells. FDA By-Lines 6:292-94 92 Geiger, L.E., and Neal, R.A. 19 81. Mutagenicity testing of 2.3.7.8-tetrachlorodibenzo-p-dioxin in histidine auxotrophs of Salmonella typhimurium. Toxicol. A p p l . Pharmacol.59:125-29 93. Reggiani, S. 1980. Acute exposure to TCDD in Seveso, Italy. J. Toxicol. Environ. Health. <5:27 94. Toth, K., Somfai-Relle, S., Sugar, J., Bence, J. 1979. Carcinogenicity testing of herbicide 2,4,5-trichlorophenoxyethanol containing dioxin and of pure dioxin in Swiss mice. Nature 278 (5704):548-9 95. DiGiovanni, J., Viaje, A., Berry, D.L., Siaga, T., Juchau, M.R. 1977. Tumor-initiating ability of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and Arochlor 1254 in the two-stage system of mouse skin carcinogenesis. Bull Environ. Contam. Toxicol. 18(5):552-57 96. VanMiller, J.P., Allen, J.R. 1977. Chronic toxicity of 2.3.7.8-tetrachlorodibenzo-p-dioxin in rats. F e d . Pro c . Fed. Am. Soc. Exp. Biol. 36(3):396 97. VanMiller, J.P., Lalich, J.J, Allen, J.R. 1977. Increaed incidence of neoplasms in rats exposed to low levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Chemosphere 6(10):625-32 156 98. Shefner, A.M. 1978. Cited in Nature 274:533-34 99. Berry, D.L., Slaga, T.J., DiGiovanni, J. 1979. Studies with chlorinated dibenzo-p-dioxins, polybrominated biphenyls and polychlorinated biphenyls in a two-stage system of mouse skin carcinogenesis; Potent anticarcinogenic effects. Ann. N.Y. Acad. Sci. 320:405-14 100. Cohen, G.M., Bracken, W.M., Radhakrishnan, P.I., Berry, D., Selkirk, J.K., Slaga, T.A. 1979. Anticarcinogenic effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin on benzo(a)pyrene tumor initiation and its relationship to DNA binding. Cancer Res. 39:4027-33 101. National Toxicology Program. 1982. Technical report of the carcinogenesis bioassay of 2,3,7,8-tetrachlorodibenzo-pdioxin in Osborne-Mendel Rats and B6C3F, mice (gavage study). NTP-80-31 NIH Pub. No. 82-1765. 102. Pitot, H.C., Godsworthy, T., Campbell, H.A., Poland, A. 1980. Quantitative evaluation of the promotion by 2,3,7,8-TCDD of hepatocarcinogenesis from diethylnitrosamine. Cancer Res. 40:3616-620 103. Poland, A. and Glover, E. 1979. An estimate of the maximum in vivo covalent binding of 2,3,7,8-tetrachlorodibenzo-pdioxin to rat liver protein, ribosomal RNA and DNA. Cancer Res. 39:3341-44 104. Reggiani, G. 1982. Toxicology of TCDD and related compounds: Observations in man. Pergamon Series Environ. S c i . 5^:463-93 105. Tognoni, G. and Bonaccorsi, A. 1982. Epidemiological prob­ lems with TCDD (A critical review). Drug Metab. Rev. 13(3): 447-69 106. Hay, A.W.M. 1982. Exposure to TCDD: The health risks. Pergamon Series Environ. S c i . 5^:589-600 107. VanNess, G.P., Solch, J.G., Taylor, M.L., Tierman, T.O. 1980. Tetrachlorodibenzo-p-dioxins in chemical wastes, aqueous effluents and soils. Chemosphere S):553-63 108. Holmstedt, Bo. 1980. Prolegomena to Seveso. Arch. Toxicol. 44:211-31 109. "Toxin Haunts River Town." The Detroit News Jan. 7, 1983 no. 138. pp. 2-A 110. Hay, A. 1977. Seveso solicitude. Nature 267(5610):384-85 111. Hay, A. 1977. Dioxin damage. Nature 266:7-8 157 112. Mahle, N . , Higgins, H., Getzendener, II. 1977. Search for the presence of 2, 3,7,8-tetrachlorodibenzo-p-dioxin in bovine milk. Bull. Environ. Contam. Toxicol. 18(2):123-30 113. Kocher, E., Mahle, N., Hummel, R., Shadoff, L. 1978. A search for the presence of 2,3,7,8-tetrachlorodibnzo-pdioxin in beef fat. Bull Environ. Contam. Toxicol. 19*229-36 114. Shadoff, L., Hummel, R., Lamparski, L., Davidson, J. 1977. A search for 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in an environment exposed annually to 2,4,5-trichloror phenoxyacetic acid ester (2,4,5-T) herbicides. Bull. Environ. Contam. Toxicol. 18(4):478-85 115. Newton, M., and Snyder, S. 1978. Exposure of forest herbi­ vores to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in areas sprayed with 2,4,5-T. Bull. Environ. Contam. Toxicol. 20:743-50 116. Woolson, E., Ensor, P., Riechel, W., Young, A. 1973. Dioxin residues in Lakeland sand and Bald Eagle samples. Advan. Chem. Ser. 120:112-13 117. Streisinger, G., 1978. Assessment of hazards posed by TCDD. In: Symposium on the Use of Herbicides in Forestry. 118. Hay, A. 1978. Vietnam's dioxin problem. Nature (London) 271(5646):597-98 119. Anonymous. 1978. Liver cancer linked to dioxin. New Scien­ tist. 78:431 120. Baughman, R . , and Meselson, M. 1973. An analytical method for detecting TCDD (dioxin): Levels of TCDD in.samples from Vietnam. Environ. Health Perspect. j>:27-36 121. Young, A.L., Thalken, C.E., Ward, W.E. 1975. Studies of the ecological impact of repetitive aerial spraying of herbi­ cides on the ecosystem of the test area C-52A, Elgin A F B , Florida. Nat'l Tech. Inform. Serv. AD-A032,773: 142pp. 122. Mieure, J.P., Hicks, 0., Kaley, R.G., Michael, P.R. 1977. Determination of trace amounts of chlorodibenzo-p-dioxins and chlorodibenzofurans in technical grade pentachlorophenol. J. Chromatogr. S c i . 15:275-77 123. Blaser, W.W., Bredeweg, R.A., Shadoff, L.A., Stehl, R.H. 1976. Determination of chlorinated dibenzo-p-dioxins in pentachlorophenol by gas chromatography-mass spectrometry. Anal. Chem. 48:984-86 124. Pfeifer, C.D. 1976. Determination of chlorinated dibenzop-dioxins in pentachlorophenol by liquid chromatography. J. Chromatogr. S c i . 14:386-91 158 125. Villanueva, E.C., Burse, V.W., Jennings, R.W. 1973. Chlorodibenzo-p-dioxin contamination of two commercially availabe pentachlorophenols. J. Agr. Food Chem. 21(4):739-40 126. Zabik, M.J. and Kaczmar, S.W. 1980 Unpublished data. 127. DaRoss, B., Merrill, R., Willard, H., Wolbach, C. 1982 Emissions and residue values from waste disposal during wood preserving. Project Summary, EPA-600/S2-82-062 128. Firestone, D., Clower, M., Borsetti, A., Teske, R., Long, P. 1979. Polychlorodibenzo-p-dioxin and pentachlorophenol residues in milk and blood of cows fed technical penta­ chlorophenol. J. A g r . Food Che m . 22(6):1171-77 129. Olie, K., Vermeulen, P.L., Hutzinger, 0. 1977. Chlorodibenzo-p-dioxins and chlorodibenzofurans are trace components of fly ash and flue gas of some municipal incinerators in the Netherlands. Chemosphere 1;455-59 130. Buser, H.R., Bosshardt, H.P., Rappe, C. 1978. Identifi­ cation of polychlorinated dibenzo-p-dioxin isomers found in fly ash. Chemosphere 2 i 165-72 131. Liberti, A. and Brocco, D. 1982. Formation of polychlorodibenzodioxins and polychlorodibenzofurans in urban incinerator'emissions. Pergamon S e r . Environ. Sc i . 5:245-51 132. Buser, H.R. and Bosshardt, H.P. 1982.-Chlorinated dibenzop-dioxins: areview of mechanisms of formation. Chemosphere 9(7-8):501-22 133. Cavallera, A., Bandi, G., Invernizzi, G., Luciani, L., Mongini, E., Gorni, A. 1980. Sampling, occurrence and evaluation of PCDDs and PCDFs from incinerated urban solid waste. Chemosphere 2(10):611-21 134. Eiceman, G.A., Clement, R.E., Karasek, F.W. 1981. Variations in concentrations of organic compounds includ­ ing polychlorinated dibenzo-p-dioxins and polynuclear . . aromatic hydrocarbons in fly ash from a municipal incin­ erator. A n a l . Ch e m . 53:955-59 135. Lustenhower, J.W., Olie, K., Hutzinger, 0. 1980. Chlorin­ ated dibenzo-p-dioxins and related compounds in inciner­ ator effluents: a review of measurements and mechanisms of formation. Chemosphere :501-22 136. Gizzi, F., Reginato, R., Benfenati, E., Fanelli, R. 1982. Polychlorinated dibenzo-p-dioxins (PCDD) and poly­ chlorinated dibenzofurans (PCDF) in emissions from an urban incinerator. 1. Average and peak values. Chemosphere 11:250-57 159 137. Cavallero, A., Luciani, L., Ceroni, G . , Rocchi, I., Invernizzi, G., Gorni, A. 1982. Summary of results of PCDD analyses from incinerator effluents. Chemosphere 11(9):859-68 138. Olie, K., Lustenhouwer, J., Hutzinger, 0. 1982. Polychlor­ inated dibenzo-p-dioxins and related compounds in incin-:.. erator effluents. Pergamon S e r . Environ. S c i . 13: 227-44 139. Hryhorczuk, D., Ulithrow, W., Hesse, C., Beasley, V. 1981. A wire reclamation incinerator as a source of environmental contamination with tetrachlorodibenzo-pdioxins and tetrachlorodibenzofurans. Arch. Environ. Health 3_6 (5) :228-34 140. Tierman, T., Taylor, M., Solch, J., Van Ness, G., Garrett, J., 1982. Analytical methodology for determination of chlorodibenzodioxins and chlorodibenzofurans in fly ash and stack effluents from municipal waste incinerators. Preprint extended abstract. A.C.S. Div. Env. Chem. 185th National Mtg. 22(1):389-94 141. Nestrick, T.J. and Lamparski, L.L. 1982. Personal commun­ ication cited in: Lustenhouwer, J., Olie, K., Hutzinger, 0. 1980. Chlorinated dibenzo-p-dioxins and related comr pounds in incinerator effluents: A review of measurements and mechanisms of formation. Chemosphere £(7-8):501-22 142. Helder, T., Stutterheim, E., Olie, K. 1982. The toxicity and toxic potential of fly ash from municipal incinerators assessed by means of a fish early life stage test. Chemo­ sphere 11(10):965-72 143. Eiceman, G. and Rghei, H. 1982. Chlorination reactions of 1,2,3,4-tetrachlorodibenzo-p-dioxin on fly ash with HC1 in air, Chemosphere 11(9):833-39 144. Lustenhouwer, J., Olie, K., Hutzinger, 0. 1980. Chlorin­ ated dibenzo-p-dioxins and related compounds in inciner­ ator effluents. A review of measurements and mechanisms of formation. Chemosphere 9(7-8):501-22 145. Liberti, A., Brocco, D., DiPalo, V., Possanzini,; M. 1978 In: Analytical Techniques in Environmental Chemistry. J. Albaiges, ed. Pergamon Press, p. 157-66 146. Stehl, R.H. and Lamparski, L.L. 1977. Combustion of several 2,4,5-trichlorophenoxy compounds, formation of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Science 19 7 (4307): 1008-1009 160 147. Rappe, C., Marklund, S., Buser, H.R., Bosshardt, H.P. 1978. Formation of polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) by burning or heating chlorophenates. Chemosphere l i 269-81 148. Jansson, B., Sundstrom, G., Ahling, B. 1978. Formation of pplychlorinated dibenzo-p-dioxins during combustion of chlorophenol formulations. Sci. Total Environ. . 10:209-17 149. Nestrick, T.J., Lamparski, L., Stehl, R. 1979. Synthesis and identification of the 22 tetrachlorodibenzo-p-dioxin isomers by high performance liquid chromatography and gas chromatography. A n a l . Che m . 5JL(13) :2273-81 150. Buser, H., Bosshardt, H., Rappe, C. 1978. Formation of polychlorinated dibenzofurans (PCDFs) from the pyrolysis of PCB's. Chemosphere 1:109-19 151. Buser, H.R., Bosshardt, H.P., Rappe, C. 1978. Identifi­ cation of polychlorinated dibenzofuran isomers in fly ash and PCB pyrolyses. Chemosphere 5^:419-29 152. Buser, H. 1979. Formation of polychlorinated dibenzo­ furans (PCDFs) and polychlorinated dibenzo-p-dioxins (PCDDs) from the pyrolysis of chlorobenzenes. Chemosphere 8:415-24 153. Ahling, B., Bjorseth, A., Lunde, G. 1978. Formation of chlorinated hydrocarbons during combustion of poly (vinyl chloride). Chemosphere 7:799-806 154. Choudry, G., Olie, K., Hutzinger, 0. 1982. Mechanisms in the themal formation of chlorinated compounds including polychlorinated dibenzo-p-dioxins. 155. Aubrey, N. and VanWazer, J. 1964. Equilibrium rearrange­ ment of perchlorocarbon compounds. J. A m e r . C h e m . S o c . 86:4380-83 156. Grant, D. 1974. The pyrolysis of chlorocarbons. J. A p p l . C h e m . Biotechnol. 24:49-58 157. Bumb, R . , Crumett, W . , Cutie, S., Gledhill, J., Hummel, R., Kagel, R., Lamparski, L., Luoma, E., Miller, D., Nestrick, T., Shadoff, L., Stehl, R., Woods, J. 1980. Trace chem­ istries of fire: A source of chlorinated dioxins. Science 210(4468):385-90 158. Mahle, N. and Whitting, L. 198 0. The formation of chlorodibenzo-p-dioxins by air oxidation and chlorination of bituminous coal. Chemosphere 9:69 3-99 161 159. Kirsbaum, I., Domburg, G., Sergeyeva 19 72. Extension of application range of gas chromatography to analysis of lignin pyrolysis products. Latv. PSR zinat. A k a d . Vestis., Khim. S e r . P. 700-07 cited in reference 154 160. Ahling, B. and Lindskog, A. 19 82. Emission of chlorinated organic substances from combustion. Pergamon S e r . Environ. Sc i . 5:215-25 161. Shenfel’d, B., Bezulklednikov, Y., Ron'shin, A., Ketov, A., Sivkova, L. 1982. Chlorination and pyrolysis of petroleum coke in a potassium chloride - sodium chloride melt. Izv. Vyssh. Uch e b . Zaved., Tsvet. M e t . 15;77-9 cited in reference 154 161a. Matsumura, F. and Benezet, H.J. 1973. Environmental generation and degradation of dibenzodioxins and dibenzo­ furans. Env. Health Perspect. J5:253-58 162. Kearney, P.C., Isensee, A.R., Helling, C.S., Woolson, E.A., Plimmer, J.R. 1973. Environmental significance of chlorodioxins. Adv. Chem. S e r . 120(11):105-11 163. diDomenico, A., Viviano, G., Zapponi, G., 1982. Environ­ mental persistence of 2,3,7,8 TCDD at Seveso. Pergamon Ser. Environ. Sci. 5:105-14 164. Philippi, M., Schmid, J., Wipf, H.K., Hutter, R. 1982. A microbial metabolite of TCDD. Experientia 38:659-61 165. Kearney, P.C., Woolson, E.A., Ellington, C.P. 1972. Persistence and metabolism of chlorodioxins in soils. Environ. S c i . Technol. 6(12):1071-19 166. Ward, C.T. and Matsumurs, F. 1978. Fate of 2,3,7,8- tetra­ chlorodibenzo-p-dioxin (TCDD) in a model aquatic environ­ ment. A r c h . Environ. Contam. Toxicol. T_CS) :349-57 167. Hutter, R. and Philippi, M. 1982. Studies on microbial metabolism of TCDD under laboratory conditions. Pergamon S e r . Environ. Sci. 5:87-93 168. Isensee, A. and Jones, G. 1975. Distribution of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) in an aquatic model ecosystem. Environ. S c i . Technol. 9^7):668-72 169. Wipf, H.K., Homberger, E., Neuner, N., Ranalder, U., Vetter, W., Vuilleumier, J. 1982. TCDD levels in soil and plant samples from the Seveso area. Pergamon S e r . Environ. Sci. 5:115-26 162 170. Isensee, A. and Jones, G. 1971. Absorption and translo­ cation of root and foliage applied 2,3-dichlorophenol, 2,7-dichlorodibenzo-p-dioxin and 2,3,7,8-tetrachlorodibenzo-p-dioxin. J. Agric. Food Chem. 19.(6) :1210-14 171. Cerquiglini Monteriolo, S., diDomenico, A., Silano, V., Viviano, G., Zapponi, G. 1982. 2,3,7,8-TCDD levels and distribution in the environment at Seveso after the Icmesa accident on July 10th, 1976. Pergamon Ser. Environ. Sci. 5:127-36 172. Crosby, D.G. and Wong, A.S. 1977. Environmental degrada­ tion of 2,3,7,8-TCDD. Science 195:1337-38 173. Kawaguchi, S. 1980. Unpublished data 174. Nestrick, T., Lamparski, L., Townsend, D. 1980. Identi­ fication of tetrachlorodibenzo-p-dioxin isomers at the 1-ng level by photolytic degradation and pattern tech­ niques. Anal. Ch e m . 52:1865-74 175. Slawinski, J., Puzyma, W., Slawinska, D. 1978. Chemiluminescence in the photooxidation of humic acids. Photochem. Photobiol. 28:75-81 176. Khan, S. and Schnitzer, M. 1978. UV irradiation of Atrazine in aqueous fulvic acid solution. J. Environ. Sci. Health B 1 3 (3):299-310 177. Chou, Shin-Shou and Eto, M. 1980. Effects of paddy water and some photosensitizers on the photolysis of the fungicide Isoprothiolane. J. Environ. Sci. Health B15(2): 135-46 178. Tanaka, F., Wien, R . , Mansagea, E. 1979. Effect of nonionic surfactants on the photochemistry of 3-(4chlorophenyl)-1,1-dimethylurea in aqueous solution. J. Agric. Food Chem. 27(4):774-79 179. Miller, G. and Zepp , R. 1979. pollutants sorbed on suspended Technol. 13(7):860-63 180. Mazer, T. and Hileman, F. 1982. The ultraviolet photo­ lysis of tetrachlorodibenzofurans. Chemosphere 1 1 (7): 651-62 181. Anonymous. Dow pesticide seen as major, if not only source of TCDD in fish. Pesticide and Toxic Chemical N e w s . Feb. 21,1979 pp.5-6 182. Crummet, W.B. 1982. Environmental chlorinated dioxins from combustion - the trace chemistries of fire hypothesis. Pergamon S e r . Environ. S c i . 5^:253-63 Photoreactivity of aquatic sediments. Environ. S c i . 163 183. Dow Chemical Company, Michigan Division 1979. Determina­ tion of tetra-, hexa-, hepta- and octachlorodibenzop-dioxins in soil, diet and particulate samples. Method ML-AM-78-63. Unpublished document, Dow Chemical U.S.A., Michigan Division, Midland,Mich. 48640. 184. Buser, H.R. 1982. High-resolution gas chromatography of the 22 tetrachlorodibenzo-p-dioxin (TCDD) isomers. Pergamon S e r . Environ. S c i . 5:15-24 185. Cowgill, U.M., Freiser, H., Gross, M.L., Rogers, L.B., Schuetzle, D. 1979. Evaluation of the analytical chemical procedures of the Dow Chemical Company for determination of polychlorinated dibenzo-p-dioxins. Unpublished report to the Dow Chemical Company 186. Lamparski, L.L., Nestrick, T.J., Stehl, R.H. 1979. Determination of part-per-trillion concentrations of 2,3,7,8-tetrachlorodibenzo-p-dioxin in fish. Ana l . C h e m . 51(9):1453-58 187. Tiernan, T.O., Taylor, M.L., Erk, S.D., Solch, J.G., Van Ness, G., Dryden, J. 1980. Dioxins: Volume I I . Analy­ tical Method For Industrial Wastes EPA-600/2-80-157 188. Johnson, R.R. 1973. Elementary Statistics, Duxbury Press North Scituate, Massachusetts 189. Miller, R.A., Norris, L.A., Hawkes, C.L. 1973. Toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in aquatic organisms. E n v . Health Perspec£. 5^:177-90 190. Hornshaw, T. 1982. Graduate research project. Department of Fisheries and Wildlife. Michigan State University, East Lansing, Michigan 191. Cordle, F. 1981. The use of epidemiology in the regula­ tion of dioxins in the food supply. Regul. Toxicol. Pharmacol. 1^379-87 192. Stevens, K.M. 19 81. Agent Orange toxicity: a quantitative perspective. Human Toxicol. 1^:31-39 APPENDIX 164 The following section outlines the methods used for quantifying the concentration of 2,3,7,8-TCDD in the fish samples and presents examples of how selected GC-MS chro­ matograms were interpreted and quantified. Quantification was based on the mass spectrometric response for ions at m/e 322 and 332 of 12C-2,3,7,8-TCDD and 13C-2,3,7,8-TCDD, respectively. The following formula was utilized: ppt (ng/kg) 2/3,7 /8-TCDD = •§-£-§ x 5 ^ T where: A = peak area of native 2,3,7,8-TCDD in the sample 13 B = peak area of added C-2,3,7,8-TCDD in the sample 12 C = peak area of C-2,3,7,8-TCDD in the standard 13 D = peak area of C-2,3,7,8-TCDD m the standard 12 E = mass of C - 2 ,3,7,8-TCDD m the standard (ng) F = mass of 13 C-2,3,7,8-TCDD in the standard (ng) G = mass of 13C-2,3,7,8-TCDD added to the sample H = mass of the fish sample This formula is expressed as two terms. The first term calculates the concentration of 2,3,7,8-TCDD in the fish sample, based on the response of the injection aliquot at m/e 322 re­ lative to the response of the most recently injected standard. The second term is used to correct for sample recovery based on the recovery of the 13 C - 2 ,3,7,8-TCDD internal standard spiked into the ground fish sample prior to digestion and cleanup. Quantification is carried out relative to the response 165 of a standard, usually 50 pg of 2,3,7,8-TCDD. Response from this standard is reestablished every fifth injection to cor­ rect for variations in mass spectrometric response caused by instrumental drift. The equation used for calculating concentrations assumes that mass specrometric response is linear over the range in which the samples to be quantitated respond. This assumption was verified by performing a linearity check at the outset of the quantification period and at the beginning of each day. Figure A-l is a standard curve of the response measured for five consecutive injections of standards of increasing con­ centration. It demonstrates that linearity was achieved and validates the assumption. for linear response was The calculated corelation coefficient 0.9 indicating acceptable linearity over the range investigated. Figures A-2 through A-ll are examples of output from the GC-MS which were used to calculate the TCDD concentrations. The first three figures are examples of output from a method blank, the 13 C - 2 ,3,7,8-TCDD surrogate and a mixed standard while the remaining figures are GC-MS specific ion chromatograms of fish sample extracts. They are presented as examples of dif­ ferent types of output and are followed by copies of the data sheets used for calculations. case in data interpretation. Each sample represents a special They have been chosen to present the range of outputs encountered in this work. Figure A-2 is the specific ion chromatogram of a procedural blank carried through the entire procedure. The signal at 166 13 C-2,3,7,8-TCDD internal standard. m/e 332 is from the Ab­ sence of a signal at ions of m/e 320, 322 and 324indicates that the sample was free of artifacts or cross contamination. The signal area of 535 units corresponds to about 0.71 ng of the spike, or a 14.1% recovery of the surrogate. This re­ covery refers to the amount of surrogate present in the in­ jection aliquot, and represents a theoretical limit of detec­ tion of 7.1 ng/kg, assuming a minimum detectable quantity of 20 picograms of 2,3,7,8-TCDD and fish sample weight of 20 grams. Figure A-3 is a GC-MS specific ion chromatogram of 1 ng of13C-2,3,7,8-TCDD. This is a 1 ul aliquot of the solution used for spiking the fishsamples dard for quantification. at a level of 1 ng of the as well as aprimary stan­ The chromatogram demonstrates that 13 C surrogate, there is no TCDD background contributed by 12 . 12 . C-2,3,7,8- . . C-2,3,7,8-TCDD impurities m the surrogate. Figure A - 4 is a mixed standard containing 50 picograms of each of the 13 C- and 12 C - 2 ,3,7,8-TCDD standard. Some background "signal is observed, especially in the region of 18 minutes at m/e 322 and 324. This background did not appear consistently in the standard, each time it was injected. Because this chromatogram was generated between runs during quantitation of samples, the peaks probably are the result of "ghosting". However, they do not show signals at all 4 ions monitored nor do they interfere with the 2,3,7,8-TCDD peaks. The ratio of the area for the m/e 320 ion to the area at m/e 322 for this standard is 0.96. This ratio is higher than expected. The 167 mean 320/322 ratio calculated for standards injected during the quantitation period was 0.79, and ranged from 0.55 to 0.96. The next set of chromatograms represents samples deter­ mined not to contain any native 2,3,7,8-TCDD. The first chromatogram, figure A - 5, is a Carp sample originating in Lake Erie. It shows an injection aliquot recovery of 73% and a theoretical limit of detection of 13 ng/kg, but does not reveal any evidence of native 12 C-2,3,7,8-TCDD. There is some contamination at ions 322 and 324, but none corresponding to the retention time of 2,3,7,8-TCDD. Also notable is the absence of any siganal at m/e 320 eluting with the impurities. This sample can be interpreted as containing no 2,3,7,8-TCDD at the stated limit of detection of 13 ng/kg. The next sample, presented as figure A-6, is somewhat more difficult to interpret. It contains very weak signals at m/e 320 and 322 and a minimal signal at m/e 332. This sig­ nal is difficult to quantify because it is near the limit of detection. The area contribution below the top of the back­ ground signals must be estimated. The area value of 50 units corresponds to 65 picograms of the surrogate standard, or a limit of detection of 76 ng/kg. The ion signals at m/e 320 and 322 suggest a trace of 2,3,7,8-TCDD. However, because a signal was not observed at m/e 324, the sample cannot be re­ ported as containing any 2,3,7,8-TCDD. The next figure (A-7) is an example of a sample which had no recovery of the internal standard and was therefore nonquantifiable. This sample was most likely the result of an 168 elution through an alumina column with improper activity or the product of a series of inefficient transfer steps. It is useless for any type of quantitative determination. The remaining examples represent samples which contained detectable residues of 2,3,7,8-TCDD. Figure A - 8 is an example of a sample with a signal at the limit of detection. The m ea­ sured area of 6 units for the ion at m/e 332 indicates a very low recovery while the response at ions 320, 322 and 324, although quantifiable, is difficult to differentiate from the background. For this reason, the sample was judged to be too poor to qualify for reporting as a positive sample. The quanti­ fied level of 2,3,7,8-TCDD would have been 738 ng/kg at a limit of detection of 638 ng/kg. Because of the high correction factor incorporated due to the low recovery (0.16%), the calcu­ lated value is of very low precision and cannot be accepted with any certainty. Figure A-9 is output fron the sample reported as containing the highest observed residue of 2,3,7,8-TCDD. in the St. Clair River. It originated Review of the output indicates a strong signal at ions of masses 320, 322 and 324 accompanied by a sig­ nal at 332 corresponding to 60 pg of surrogate, or a theoretical limit of detction of 81 ng/kg. The ions show acceptable response by all criteria established for quantification. isotope ratio is a little high. The 320/322 However this is due to the way the 322 ion peak was integrated. The baseline for the 320 ion was determined to lie directly along the bottom of the leading and tailing edges of the peak, while the 322 ion baseline 169 was interpreted to exist above the tangent point of the base­ line. It is likely that most of the area above the baseline is due to TCDD, and could be requantified as such. This would result in a lower 320/322 isotope ratio and a higher calculated TCDD concentration. Figure A-10 is output from a sample of fish originating in the Tittabawassee River. It represents a sample quantified near the mass spectrometric limit of detection. m/e 322 corresponds to 33 pg of 2,3,7,8-TCDD. The area at Corrected for surrogate recovery of 2.6%, the concentration of 2,3,7,8-TCDD in the fish sample sample was calculated as 83 ng/kg. The last figure (A-ll) is output from a sample showing a good recovery and strong signals at all ions representing native 2,3,7,8-TCDD. This sample originated in the Muskegon River. This type of output was the easiest to interpret and holds the greatest degree of precision. 600- - 400- ■ 300- - 170 AT M/a 322 500 200- - 100- - .05 .20 .30 .40 ng 2,3,7,8 - TCDD Figure A-l GC-?1S Standard Curve of 2, 3, 7 ,8-TCDD. .50 FPM s n . i . c ve:d tom chrohhtoophh ** 1r 'Y F1.ANK lO H- ■ .25 42. A- 535. 332.0: 324.0 322.0. L20. Ol -lfi. Figure A-2 T* -12. T" _ia_ i -ia_ SIM GC-MS Chromatogram of Procedural Blank. 6152 1 1.0(1 M t f l l .F X r E D I'JH CHPOFIh TOGPMI * * 1.0 C-13 s t ^ n d p p d 1.0 NC; AS PROOIDEP H« 69. FPN 6171 ■•.spij'. 1 o '/« 1 . 0 0 703 Figure A-3 SIM GC-MS Chromatogram of 1 ng of ^ C - 2 ,3, 7 ,8-TCDD O .0 F NG EACH C - i e M StIT.RCTc.D LON CHRONmTOGPAM Jf* C - 1 3 TCDD STANDARD FPU 6011 1 C>T 'y. . ' P C : 1 .2^ V* 1.00 H= 3. H« 3. 5. H- 33. JT n T T i m w m m 3 3 2 .0 j i n n Jl 324-01 A _____________ 322.0 A- lU 67. iL 320.0 _ Ifi. Figure A-4 lE - 18 - SIM GC-MS Chromatogram of 50 pg Mixed Standard. SFI.F.crF.D ION r.Hlt’Of'liiTOGPWI ** FM1 5163 8-E 3 3 2 .0 12 Figure A - 5 SIM GC-MS Chromatogram of Sample with No Detectable 2 ,3 ,1 , 8-TCDD 175 s e SAMPLE I.D. LOCATION SPECIES WEIGHT (kg) I . H 2 - ____________ ^ V. £____________ LENGTH (cm) SOURCE IVS^U^ tp o o SAMPLE WEIGHT % FAT _____________ _______ H ________________ % MOISTURE ________ AGE FRN ______ CARTRIDGE RESPONSE 332 £??s 320 322 ~~~______ ISOTOPE RATIO 324 STANDARD RESPONSE mass S ° 332 322 »• % RECOVERY LMIT OF DETECTION CONCENTRATION / * \ _ |0 . ^ M ~t" LEG TED ION CHROi'WluGRei'l ** .2? V= STIFLE SO pr w llll I 1.00 m 332.0 MU Fr-ri 6082 lf>T -'.-pr,; | A I 176 324.0 n rrn r waMmju^Ksiijhi 322. C in i i i m i r n 320. G JLfi. Figure A-6 T “ -12. T _LS_ — r “I -1SL SIM GC-MS Chromatogram of Sample with No Detectable 2,3,7,8-TCDD 177 m SAMPLE I.D. &k LOCATION f k ) ‘ c ^ U c J L ^ . ------- SPECIES /, O ^ P WEIGHT (kg) --------- LENGTH (cm) _ J t l ± ------------SOURCE ^ SAMPLE WEIGHT % FAT U M - ________ S jO 0 ______ 4 % MOISTURE . XOJL ^ _ AGE CARTRIDGE RESPONSE 332 S'*? 320 322 _____________ 324 ^ ISOTOPE RATIO STANDARD RESPONSE So mass 332 _ 3 S 322 7. RECOVERY \ lo o r * y LMIT OF DETECTION CONCENTRATION / fyO - - -76.7 * /- 3 M - illE C -rc ID ION CHROi’I A T 'J G R e n ** so _£ SO ilT r FPN 6082 2 ST • . . . P O : 1 v* V= 1 . 0 0 HnlliUi 3 3 2 .0 mu nui i ruin n 178 3 2 4 .0 iii jMrniujl,|4jiijA iiihiimi i i.i i n 3 2 2 .0 3 2 0 .0 ■ - ■■ i _____ li S___________L? Figure A-7. ■ i .........18 ... - i _______ 12 ----------------------------------------------------- SIM GC-MS Chromatogram of Sample v/ith No Quantifiable Recovery F&U 2-4S 5 12 3 ir/r ' h ,; 1 .<■ .25 v= l.Oti H« 1. H- 1. H- 2. H* 2. 5. 25. 332. C I L L M J i n A- 28. 1 | IT T lU IH f l^ ^ l l l H l l l M l i l i l l l l Hi l l JfL Figure A-8. JJL 18 Example of Sample at the Limit of Detection 180 J?Y2 SAMPLE I.D. ~ ‘S a & J s M A f* LOCATION SPECIES ^ y . 4^p ___________ WEIGHT (kg) LENGTH (cm) _ W . f 'b P H - SOURCE HQl SAMPLE WEIGHT _ % FAT S £ ______< T % MOISTURE 7 3,3 AGE FRN CARTRIDGE RESPONSE 332 320 322 2% ISOTOPE RATIO Q -8 < l 324 STANDARD RESPONSE mass 332 _ 322 % RECOVERY 3 b Z SAMPLE WEIGHT % FAT - f o J Z O o h ^ ___________ 3- S" % MOISTURE i AGE frn __ CARTRIDGE RESPONSE 332 £L 320 / 6 6 ______ 322 / ? ? - 324 ISOTOPE RATIO d . X $ 332 _ 322 6/ % RECOVERY LMIT OF DETECTION / O O _ ^ CONCENTRATION ~ (j, o 5^1 q y 1 t VLI.tv Tr.O I PMPOUATOGRAI'l 4 * £01 H* 5. H* 2. H« 3. H* 1. t t .. - r ^ ^ n f r in n n ir ijy u ^ 332.0 A- ___ u i J H* FPU 5143 l'Vr -P-i: 1 y-. r-;n v-= l .on 2. 9. ■ n w iL n n u B jy iiL n r ^ — \ r r rL » r J ^ A- '36. 324.0 . . . . .... i A- 322 -0 ____M .JA.JH .. 1 Ju MA 31. - ..... _ i r T 320.0 ■ „ll3 Figure A-10. 1? IS SIM GC-MS Chromatogram of Sample Quantified at the Limit of Detection 184 £o[ SAMPLE I.D. ________ LOCATION SPECIES ---------- - WEIGHT (kg) ._____ 0 - ------~ ---- ------ LENGTH (cm) % .t£ . SOURCE SAMPLE WEIGHT 3- A % FAT 3 0 % MOISTURE AGE 2_ 57 ¥ 9 FRN 9 CARTRIDGE RESPONSE 332 >6 SPECIES q P M ^t HjuJ L s ^ f t.R Q WEIGHT (kg) % -3 LENGTH (cm) SOURCE Z o SAMPLE WEIGHT .. a± % FAT % MOISTURE _ S ’S - . / AGE S' FEN S'! CARTRIDGE RESPONSE 332 320 /(.S’ 322 X U ISOTOPE RATIO _ 91- 324 STANDARD RESPONSE mass __ 332 3 £ 322 & ( % RECOVERY a 13 LMIT OF DETECTION ■ ^5 " CONCENTRATION Z / ? " 0 - 9 9 -